TL;DR
This paper introduces a novel non-contact method for imaging internal defects in cylindrical components using laser-induced ultrasound and a polar coordinate-based synthetic aperture focusing technique (LU-pSAFT). The developed system achieves high-resolution imaging of small defects, demonstrating less than 1° circumferential resolution and sub-millimeter dimensional and positional accuracy.
What Problem Does This Paper Address?
Cylindrical components are ubiquitous in industrial applications, and their structural integrity is paramount for performance and safety. Internal defects, even small ones, can act as stress concentrators, leading to catastrophic failures. Traditional non-destructive testing (NDT) methods like penetrant testing (PT) and magnetic particle testing (MT) are limited to surface defects. Eddy current testing (ET) and radiographic testing (RT) can detect subsurface defects but have limitations: ET struggles with deeper defects, and RT involves harmful radiation and high costs. Ultrasonic testing (UT) offers a promising alternative due to its sensitivity and penetrating ability. However, applying UT to curved surfaces like those of cylindrical components presents challenges in achieving high-resolution imaging and accurate defect localization. Existing UT methods, including B-scan and C-scan, often lack sufficient resolution or struggle with complex geometries. Synthetic Aperture Focusing Technology (SAFT) has shown promise for high-resolution imaging, but its adaptation to cylindrical geometries requires specialized algorithms. This paper aims to bridge this gap by developing a non-contact, high-resolution imaging method specifically for internal defects in cylindrical components.
Methods
The core of this research lies in the development of a Laser Ultrasonic Synthetic Aperture Focusing Technology in the polar coordinate system (LU-pSAFT). Unlike traditional SAFT, which is often applied to flat surfaces, this method adapts the algorithm to the curved geometry of cylindrical components.
Here's a breakdown of the key methodological aspects:
- Laser Ultrasonic (LU) Excitation and Detection: The system uses a pulsed laser to generate ultrasonic waves within the material. These waves propagate through the component and reflect off internal defects. Detection is achieved using a two-wave mixing (TWM) interferometer, which is robust to rough surfaces and provides a good signal-to-noise ratio.
- Polar Coordinate System Adaptation: The standard SAFT algorithm, which relies on Cartesian coordinates, is adapted to a polar coordinate system. This is crucial for handling the circular nature of cylindrical components. The algorithm accounts for the varying distances and angles between the ultrasonic source, the defect, and the detection points in this polar framework.
- Signal Processing:
- Filtering: The raw laser ultrasonic signals contain various wave modes and noise. A band-pass Butterworth filter is employed to isolate the desired longitudinal wave signals. The paper investigates the optimal frequency range for filtering, finding that a 25–35 MHz range yields the best results for the tested aluminum alloy.
- Windowing: Due to the strong initial signal from the laser pulse and thermal expansion, a windowing technique is applied to suppress these early, high-amplitude signals that can obscure defect echoes.
- Signal Synthesis: The SAFT algorithm synthesizes a high-resolution image by coherently combining the signals received from multiple excitation and detection points. The LU-pSAFT algorithm calculates the propagation time of the ultrasonic waves in the polar coordinate system to accurately reconstruct the defect's location and shape.
- Parameter Optimization: The study investigates the influence of key parameters on imaging quality, including:
- Ultrasonic Beam Angle (β): The paper explores how the angle of the ultrasonic beam affects imaging. It's found that an angle around 100° provides the best balance, avoiding excessive signal loss from small angles and maintaining a good signal-to-noise ratio.
- Array Spacing (D): The spacing between consecutive scanning points (effectively the array spacing in the synthesized aperture) is analyzed. A smaller spacing (D = 1/4) leads to higher image quality and reduced noise, though it increases acquisition time.
Datasets
The experiments were conducted using a 1060 aluminum alloy rod with a diameter of 20 mm. This rod was intentionally fabricated with four artificial holes of 0.5 mm diameter. The locations of these holes were precisely defined by their circumferential intervals (30°) and radial intervals (0.5 mm). This sample serves as a benchmark for evaluating the performance of the developed LU-pSAFT imaging method. No other datasets were used for training or evaluation.
Key Insights and Ideas
- Polar SAFT is Key for Cylindrical Geometries: The most significant insight is the successful adaptation of SAFT to a polar coordinate system. This is not a trivial extension; it requires careful consideration of how the synthetic aperture is formed on a curved surface, moving beyond the standard Cartesian assumptions.
- Signal Filtering is Crucial for Weak Echoes: The paper highlights that the echoes from small internal defects are significantly weaker than other signal components (like Lamb waves or initial laser-induced signals). This underscores the necessity of sophisticated signal processing, particularly precise band-pass filtering and windowing, to isolate and amplify these faint defect signals.
- The "Sweet Spot" for Beam Angle: The research reveals that there isn't a single "best" ultrasonic beam angle for all scenarios. Instead, there's an optimal range (around 100° in this study) that balances signal strength, coverage, and the ability to detect subtle defect echoes without being overwhelmed by noise or losing information.
- Trade-off Between Resolution and Efficiency: The analysis of array spacing demonstrates a classic engineering trade-off. Denser sampling (smaller spacing) yields better image quality and resolution but at the cost of significantly longer data acquisition times. This necessitates finding an optimal balance for practical applications.
Key Results
The LU-pSAFT method demonstrated impressive performance in imaging the 0.5 mm diameter holes in the aluminum alloy rod:
- Accurate Defect Outline: The method successfully presented the outline of the small holes.
- High Circumferential Resolution: The circumferential resolution achieved was less than 1°.
- Excellent Dimensional and Positional Accuracy: The dimensional accuracy and position error were less than 0.1 mm.
- Improved Image Quality: Compared to traditional B-scan imaging, the LU-pSAFT method provided significantly clearer and more accurate defect localization and shape representation.
- Optimal Filtering: A 25–35 MHz band-pass filter, applied thrice with windowing, yielded the highest image quality (SNR).
Limitations and Future Work
- Sample Material and Defect Type: The study focused on a specific material (1060 aluminum alloy) and defect type (small holes). The performance of the method on other materials with different acoustic properties or on more complex defect geometries (e.g., cracks, inclusions) needs further investigation.
- Acquisition Time: While high resolution was achieved, the point-by-point scanning process for LU-pSAFT can be time-consuming, especially with very small array spacings.
- Laser Ablation: The pulsed laser can cause slight ablation on the sample surface, requiring post-processing (gentle grinding). While this is a known aspect of LU, minimizing it or developing alternative excitation methods could be beneficial.
Future work could involve:
- Extending to More Complex Geometries and Materials: Testing the method on a wider range of cylindrical components and materials.
- Investigating Different Defect Types: Evaluating the effectiveness for detecting cracks, voids, and other common manufacturing defects.
- Optimizing Acquisition Speed: Exploring techniques to accelerate the scanning process without compromising image quality, perhaps through parallel acquisition or advanced beamforming.
- 3D Imaging: Developing extensions of the LU-pSAFT method for full 3D reconstruction of internal defects.
Summary updated