The technology for inspecting defects in electrical wiring is considered highly important across various industries, including aviation, automotive, and manufacturing. Electrical wiring plays a critical role in numerous industries and everyday life due to its high conductivity and flexibility, making it adaptable to various environmental conditions. However, external environmental factors can damage wiring, degrading electrical performance and increasing the risk of fires or system failures. Therefore, industries relying on electrical wiring must implement defect detection technologies to ensure safety, efficiency, and reliability[1].

The methods for detecting defects in electrical wiring can be classified into two primary types. Contact-based methods, such as time domain reflectometry (TDR), analyze electrical characteristics like resistance and impedance to identify issues[2, 3]. The TDR methods have been widely utilized to detect and locate faults in power electrical line systems. However, they often are suitable for offline detection since they need to detect the relatively large pulse signals that would affect the original system[3]. In contrast, non-contact methods that use optical, ultrasonic, and X-ray imaging technologies detect defects in electrical wiring by analyzing structural inhomogeneities in the visualized images[4-6]. Because non-contact imaging methods do not disturb the electrical system under test, they can be applied to detect defects in electrical wiring in online systems. Especially, imaging methods can directly obtain the location of defects from measurement images, so they do not require complex analyses like TDR methods to find the location of defects. Although non-contact imaging methods have many advantages over contact-based TDR methods, such as high spatial resolution and high-speed measurement throughput, they have limitations due to their measurement principles. For example, optical methods are limited to optically transparent materials, ultrasonic imaging requires a contact medium and struggles with rough surfaces and complex structures, while X-ray imaging demands expensive equipment and raises concerns about radiation safety[4-6].

In a previous study, we demonstrated that a conductive defect embedded in an optically opaque dielectric material can be detected using microwave near-field imaging[7, 8]. Because microwaves penetrate well into dielectric materials, they can interact with conductive objects within these materials, leading to changes in the microwave near-field structure[7]. Therefore, it is expected that by visualizing the microwave near-field distribution around electrical wiring embedded in dielectric material, electrical defects can be detected from changes in the microwave near-field structure without requiring electrical contact.

In this study, we report a new non-contact method for detecting electrical faults in wires using microwave near-field imaging with thermo-elastic optical indicator microscopy (TEOIM)[7-11]. Various types and locations of faults were introduced into the wires under investigation, and the changes in the microwave near-field distribution caused by the fault conditions of the wires were investigated. From the experimental results, we confirmed that electrical faults in wires can be detected in a non-contact and non-destructive manner through the microwave near-field distribution imaging.

Figure 1 presents the TEOIM measurement setup, which is divided into two subsystems: one for microwave signal generation and application, and the other for optical signal detection. The microwave subsystem comprises a signal generator that emits microwave signals at 10 GHz, an amplifier that increases the signal strength to 30 dBm, and a TE-mode open-ended waveguide antenna that directs the microwaves onto the sample. This waveguide antenna can rotate, allowing control over the orientation of the microwave electric field with respect to the device under test (DUT). The optical subsystem is similar to a traditional polarized light microscope and includes a light source (light emitting diode; wavelength: 530 nm), circular and linear polarizers, an optical indicator, a zoom lens, and a CCD camera. During the measurement process, light from the source is first converted into left-circular polarization by the circular polarizer. The polarized light then travels through the optical indicator, reflects, and is further filtered through the linear polarizer. Finally, the intensity of the resulting transmitted light is captured by the CCD camera, which is connected to the zoom lens for detailed observation.

Figure 1. (Color online) Measurement system of TEOIM: A microwave signal generated by a signal generator was amplified to 30 dBm by an amplifier, and the amplified signal was transmitted to a TE-mode open-ended waveguide antenna. The waveguide antenna emitted microwaves to the device under test (DUT). The optical system consisted of a light source (wavelength: 530 nm), circular and linear polarizers, an optical indicator, a zoom lens, and a CCD camera. The probing light from the light source was first converted into left-circular polarization by the circular polarizer. The polarized light then passed through the optical indicator, reflected, and was further filtered through the linear polarizer. Finally, the intensity of the transmitted light was captured by the CCD camera.

For this study, we utilized a soda-lime glass plate (100 mm × 100 mm × 1 mm) coated with a 200 nm thick layer of Indium Tin Oxide (ITO, sheet resistivity < 10 ohm/sq) as the optical indicator. To mitigate the effects of reflected light from the sample, the optical indicator was mounted on an alumina substrate (1 mm thick) that is optically opaque. Conductive wires were attached to the alumina substrate, and electrical defects in the wires were created by cutting them. The length and diameter of the conductive wire were 20 cm and 0.3 mm, respectively. The optical indicator attached with conductive wires were placed 5 cm away from the waveguide antenna.

When microwaves are applied to the sample, a microwave near-field is created through the interaction between the microwaves and the sample, varying according to the electromagnetic properties and spatial configuration of the sample. If the optical indicator is placed near the sample, its microwave-absorbing layer converts the microwave magnetic field into heat, which is then transferred to the glass substrate of the indicator[9]. Consequently, the heat source distribution in the substrate reflects the distribution of the magnetic microwave near-field. The heat source distribution on the glass substrate can be visualized by analyzing the thermal stress distribution, which is measured through the photo-elastic effect of the substrate, as described by the following equation[9]:

where, q is the heat source density of in the indicator, x and y are the spatial coordinates of the image pixels of CCD sensor, and C is a constant dependent on the wavelength of incident light and the optical properties of the indicator. β1 and β2 in Eq. (1) are the linear birefringence coefficients induced by normal and shear stress in the sample. They can be calculated from the intensity changes measured by the CCD camera using the following equation[9]:

where, I is the light intensity, θ is the angle between the principal axis of thermal stress and the direction perpendicular to the horizontal plane, φ is the angle between the linear polarizer and the direction perpendicular to the horizontal plane, MW_ON and MW_OFF are the states with and without microwave application, respectively.

To investigate the interaction between a conductive wire and a microwave electromagnetic field, we visualized the microwave near-field distribution around a conductive wire excited by a microwave polarized both parallel and perpendicular to its longitudinal direction. Figures 2(a–b) and 2(c–d) show illustrations of the measurement setup and the visualized microwave magnetic near-field (H-MWNF) structures measured under parallel and perpendicular configurations, respectively. From the measurement results, it can be seen that when the microwave electric field direction is parallel to the longitudinal direction of the wire, a strong H-MWNF appears on the conductive wire. However, in the perpendicular configuration, there is no significant H-MWNF intensity on the conductive wire. These measurement results indicate that the conductive wire interacts strongly with the applied microwave when the microwave electric field is parallel to the longitudinal direction of the wire. This direction-sensitive interaction of the conductive wire with the microwave electric field can be understood in terms of the anisotropic polarizability of the wire. Since the conductive wire can be polarized more effectively along its length than in its diameter direction, the coupling between the conductive wire and the microwave increases when the microwave electric field is parallel to the length of the wire, leading to more pronounced microwave excitation of the wire.

Figure 2. (Color online) (a) Illustration of the parallel configuration for a defect-free wire, where the microwave electric (MWE) field is parallel to the length of the wire. (b) H-MWNF distribution of the defect-free wire measured in the parallel configuration. (c) Illustration of the perpendicular configuration for a defect-free wire, where the MWE field is perpendicular to the length of the wire. (d) H-MWNF distribution of the defect-free wire measured in the perpendicular configuration. (e) Illustration of the parallel configuration for a defected wire, where the MWE field is parallel to the length of the wire. (f) H-MWNF distribution of the defected wire measured in the parallel configuration. (g) Illustration of the perpendicular configuration for a defected wire, where the MWE field is perpendicular to the length of the wire. (h) H-MWNF distribution of the defected wire measured in the perpendicular configuration.

To explore the change in the microwave near-field distribution when there is an electrical defect in the wire, we visualized the microwave near-field distribution of a wire with an electrically broken section. Figures 2(e–f) and (g–h) show illustrations of the measurement setup and the visualized H-MWNF structures measured under parallel and perpendicular configurations, respectively. From the H-MWNF distributions measured under the parallel configuration, it can be seen that regions with strong H-MWNF appear periodically along the longitudinal direction of the wire. These H-MWNF distributions imply that a standing wave was created along the conductive wire, and this was caused by a discontinuity in electrical conductivity at the broken section of the conductive wire. Beside the standing wave pattern of H-MWNF, it was observed that a significant decrease in H-MWNF intensity around the broken section of the conductive wire. This result can be explained by the fact that the H-MWNF on the conductive wire is generated by the microwave current flowing through it. Because microwave current cannot flow through the broken section, the H-MWNF does not appear at the broken section. These measurement results indicate that the occurrence of the standing wave pattern and a significant decrease in the intensity of the H-MWNF structure can be a fingerprint for detecting the presence of an electrical fault in the conductive wire.

While the H-MWNF distributions measured under the parallel configuration showed interesting structures that provided information on the presence of an electrical fault in the conductive wire, the H-MWNF distributions measured under the perpendicular configuration showed only subtle changes in the H-MWNF structure. This can be explained by the fact that, because the coupling between the conductive wire and the microwave under the perpendicular configuration is weak, the microwave current is barely excited in the wire, and thus the H-MWNF is not formed. These results indicate that, for detecting electrical faults in a conductive wire, the polarization direction of the microwave electric field must be aligned parallel to the longitudinal direction of the conductive wire.

Finally, we investigated the changes in the microwave near-field that occur when there is an electrical fault in one of two adjacent conductive wires. Figure 3(a–b) and (c–d) show illustrations of the measurement setup and the visualized H-MWNF structures measured under parallel and perpendicular configurations, respectively. From the measurement results, it can be seen that strong H-MWNF appears on the two conductive wires in the parallel configuration, while there is no significant H-MWNF intensity on the conductive wires in the perpendicular configuration. These results indicate that the microwave currents in the conductive wires were efficiently excited in the parallel configuration, even when the conductive wires were placed close to each other. Figure 3(e–f) and (g–h) show illustrations and measured H-MWNF structures of two conductive wires, where an electrical defect was created in the left-side conductive wire. From the measurement results, it can be seen that a standing wave structure of the H-MWNF occurred in both the defect-free and defected conductive wires. The occurrence of the standing wave structure in the defect-free conductive wire suggests that there is microwave coupling between the defect-free and defected conductive wires, resulting in the excitation of a standing wave mode in the defect-free conductive wire. In addition, for the defective conductive wire, there is a significant decrease in H-MWNF intensity at the defective section, while the defect-free conductive wire shows significant H-MWNF intensity throughout. These results indicate that, to determine the presence of an electrical defect in a conductive wire, both the formation of a standing wave and a significant decrease in H-MWNF must be considered simultaneously. This can be confirmed by experimental results for two defected conductive wires. Figure 3(i–j) and (k–l) show illustrations and measured H-MWNF structures for two conductive wires, where electrical defects were created in both wires at different positions. From the measurement results, it can be seen that the standing wave patterns in the two conductive wires were more pronounced compared to the previous results. In addition, for both defective sections of the conductive wires, there was a significant decrease in H-MWNF intensity. These results clearly indicate that when an electrical defect is created in the conductive wire, a significant decrease in H-MWNF intensity occurs at the defective section of the wire. Therefore, it can be concluded that by visualizing the H-MWNF distribution of a conductive wire, one can determine the presence and position of an electrical defect based on the formation of a standing wave pattern and a localized significant decrease in H-MWNF intensity.

Figure 3. (Color online) (a–b) Illustration of the parallel configuration for two defect-free wires and the measured H-MWNF distribution. (c–d) Illustration of the perpendicular configuration for two defect-free wires and the measured H-MWNF distribution. (e–f) Illustration of the parallel configuration for defect-free and defected wires and the measured H-MWNF distribution. (g–h) Illustration of the perpendicular configuration for defect-free and defected wires and the measured H-MWNF distribution. (i–j) Illustration of the parallel configuration for two defected wires and the measured H-MWNF distribution. (k–l) Illustration of the perpendicular configuration for two defected wires and the measured H-MWNF distribution.

In conclusion, we report a new non-contact method for detecting electrical faults in wires using microwave near-field imaging with thermo-elastic optical indicator microscopy. We investigated the change in the microwave near-field distribution of a conductive wire caused by an electrical defect in the wire. The measurement results showed that when an electrically broken section is created in the conductive wire, a standing wave pattern appears along with a significant decrease in microwave near-field intensity at the broken section. Based on the experimental results, we showed that the presence and location of an electrical defect in a conductive wire can be determined by visualizing the microwave near-field distribution and observing the occurrence of a standing wave pattern and a significant decrease in microwave near-field intensity.

This work was supported by the 2024 education, research and student guidance grant funded by Jeju National University.