Optical defect inspection is particularly beneficial for industrial use, as it can measure microscopic and structural defects of various materials. This technique can be applied to the measurement of micro-objects (electronic composite structures, etc.) in high-technology fields such as aerospace, automobile, and novel materials, which enables the nondestructive inspection of various machinery and electrical/electronic products.
As the conventional optical defect inspection systems use a single laser or dual lasers to form a measurement region, the measurement region for defect detection is narrow.
This study attempted to overcome this disadvantage by applying multiple laser diodes and controlling the power such that the measurement region could be expanded.
Herein, chapter 2 explains the optical defect inspection system using multiple laser diodes. Chapter 3 describes the experiment for measuring the defects in the pressure piping using the proposed optical defect inspection system. Chapter 4 presents the analysis of the experimental results, and finally chapter 5 presents the discussion.
In shear interferometry, a speckle correlation fringe pattern is formed by connecting phase points in the degree of the speckle wavelength. The change in the speckle pattern according to object deformation can be observed in real time by performing a consecutive cut process for the nondeformed speckle pattern using an image processor [1].
Fig. 1 is a schematic showing the principle of the shear interferometer. The fringe analysis using the shear interferometer expresses the deformations of two points as the difference between d and d′, as shown in Eq. (1). Here, the deformations are caused by the shearing amount when a beam is irradiated on point P. n0 is the irradiation direction and ns is the vector of the observational direction [1].
If the angle of beam illumination is perpendicular to the imaging plane, then Eq. (1) can be expressed as Eq. (2) [1].
Δd can be expressed by Eq. (3) using the Taylor series. The deformation in the x-direction, Δx is equal to the shearing amount, s on the imaging plane. As this value is small, terms with two or more degrees can be ignored in Eq. (3). Eq. (2) can be expressed as the following Eq. (4) [1]:
The light source from a laser was irradiated and diffused on a specimen through the beam expander. The light reflected from the object surface was divided into two beams by a beam splitter (BS). These beams entered the charge-coupled device (CCD) camera again through piezoelectric (PZT) mirror 1 and mirror 2, respectively. At this time, the light source with changed phase entered PZT mirror 1 and the sheared light source entered mirror 2. When these two light sources met on the imaging plane, they interfered with each other. The phase change due to shearing can be expressed by the following equations. The phase difference at P is described by Eq. (5) [1].
Eq. (6) shows the phase difference at Q [1].
As the interferometer measured the difference in deformation between two points P and Q, the phase difference Δϕ, where a fringe pattern occurred in the interferometer, was obtained by subtracting Eq. (6) from Eq. (5), which can be expressed by Eq. (7) [1]:
In addition, the phase differences at P and Q between before and after deformation can be directly calculated, as shown in Eqs. (8) and (9), respectively.
In the equations above (1-7) 
, and 
are normal lines to each other. Using Taylor’s theorem, Δd can be developed as follows:
Since Δx is equal to Sx on the imaging plane and this value is very small, Eq. (10) can be expressed as Eq. (11) [1]:
Accordingly, when the values of Δd1 and Δd2 are obtained from Eqs. (1-11) and substituted into Eq. (9), the following Eq. (12) is derived:
Additionally, if θ is a small value, sin θ is zero. Accordingly, Eq. (13) is obtained.
Consequently, Eq. (14) can be derived. As is clear from Eq. (14), the shearography could directly measure the differential values of the out-of-plane deformation in the case of a small shearing amount. As the shearing amount ∂x decreases, the function becomes closer to a derivative and the sensitivity (relative displacement) of an optical system can be adjusted by the shearing amounts. Shearography or speckle pattern shearing interferometry measures the relative displacement between two points that are as distant as the shearing amount of the out-of-plane displacement [1].
In Eq. (14), if the shearing amount shows lim ∂x → 0, the measurement becomes close to the first derivation. The measurements of shearography are typically expressed as first derivatives. However, any shearography that does not satisfy the differential condition (lim δx→0) measures the relative displacement between two points not by the first derivatives but according to the shearing amounts [1].
The multiple diode optical defect inspection system is a noncontact type system using multiple laser diodes as shown in Fig. 2. This system measures shapes to detect the out-of-plane deformation of defects. As the system integrates the shearing interferometer-based optical measurement system, an analysis and image processing program, and a control program, the internal defects and shapes of pressure vessels and piping can be measured.
As shown in Fig. 3, the proposed system consists of a lens set, a CCD camera, a beam splitter that divides light, mirror 1 connected to a PZT actuator for precision transport, and the tilting mirror 2 that applies shearing angles. To use six laser diodes as the light source, a laser diode arm fixture was designed on the right plate such that the object could be illuminated with diffused reflection.
The conventional single light source diffuses light through an expanding lens. Accordingly, the measurement region is determined by the diffusion range, which is again based on the lens curvature, focal length, and material. When the diffusion range increases within the same distance, the light intensity decreases, which reduces the measurement region. To solve this problem, the laser power can be increased. However, this solution increases the cost. As a laser diode is cost effective and has the same specifications, using multiple number of laser diodes can expand the diffusion range and the light intensity can also be compensated for the expansion. Consequently, the measurement region can be effectively increased, as compared to the conventional method.
Two types of specimens for the pressure pipe were fabricated by applying the defect angle of 90° and arbitrary defects. To model the pressure pipes of a nuclear power plant, the specimens had the radius of curvature of 250 mm in the circumferential direction, which corresponded to the tool diameter. The Shape and dimension of piping test specimens are shown in Table 1.
| Description | Specimen data |
|---|---|
| Length of pipe (L) | 400 mm |
| Outside diameter (Do) | 76.4 mm |
| Inside diameter (Di) | 69 mm |
| No Defect thickness (tp) | 3.7 mm |
The pipe defects in nuclear power plants were reproduced by welding an intact part and a defective part, which included internal defects under the conditions of Table 2, to fabricate a single pipe specimen. The goal of fabricating the specimens with such defects was to measure the internal thinning defects that occurred as the complex effect of erosion and corrosion when the circulation rate increased [7].
Fig. 4 shows the shape and dimension of the specimen. The specimen material is the ASTM A53 Gr.B type F, its yield strength is 240 MPa, tensile strength is 415 MPa, and its maximum longitudinal elongation percentage is 30 %. The depth, width, and length of the defects, which were measured in the experiment, are also presented in each condition.
A modeling system was fabricated to reproduce a defective pressure vessel used in a nuclear power plant or an industrial plant. Fig. 5 illustrates the modeled image of the pipe circulation system, and Fig. 6 shows the real image of the system.
In the experiment, an enclosed pressure vessel was heated using electric coils and steam was generated inside the vessel. The generated steam circulated along the internal pipe and nuclear plant pipes and then returned to the vessel. The maximum pressure in the pressure vessel was 0.2 MPa, and the highest temperature was 125-130 °C. A safety device was also installed to maintain the internal pressure of the vessel under a high temperature and pressure. The system temperature can be controlled by a control box.
Tables 3 to 6 show the phase maps and line profiles above were obtained using the multiple diodes optical defect inspection system. The phase maps and line profiles above were obtained using the multiple diodes optical defect inspection system. The phase maps were based on variations in the internal temperature of the specimens, and the defect width and length. The line profiles show the measurements of deformation in the defective parts.
| No. 01/02 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 03/04 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 01/02 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 03/04 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 01/02 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 03/04 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 01/02 | Phase map |
 |
|
| Line profile |
 |
 |
|
| No. 03/04 | Phase map |
 |
|
| Line profile |
 |
 |
|
In the experiment, the phase maps of the defects could be identified according to the temperature variations inside a pressure pipe using the inspection system. We found that the defects could be measured according to the defect width and length using the line profiles. The detectable defects ranged from approximately 0.05 μm to 0.40 μm. The detection results were not significantly different from those obtained using a single laser; instead, defects could be detected over a wider range.
An optical defect inspection system using multiple laser diodes was used to measure specimen defects caused by external factors. The internal defects could be measured with a measurement resolution of 1 μm, and the out-of-plane deformation was measured with a system resolution of 0.1 μm. The conventional laser interferometers have different measurement regions depending on the CCD or CMOS specifications and the performance of the zoom lens. However, this study showed that the measurement region of approximately 200 mm × 100 mm can be obtained using a single laser but the maximum measurement region of approximately 500 mm × 400 mm can be secured by applying multiple lasers under the same condition.
If the proposed optical defect inspection system using multiple laser diodes is adopted in real industrial sites, such cost-effective laser diodes will produce the equivalent performance to that of expensive lasers, thereby enabling the low-cost and high-efficiency nondestructive inspection.
Moreover, as the detection range for each measurement is widened, the measurement duration is expected to be reduced. Consequently, the proposed inspection system will be particularly beneficial for large-scale plants, wind power stations, aerospace materials, and any other fields demanding nondestructive inspections of a wide area.
