Abstract : This paper studies the array testing technology of eddy current sensors for real-time position monitoring of large-area curved metal components using a flat flexible eddy current sensor array. A time-division multiplexing-based eddy current array testing method is adopted. Through rational design of the sensor probe and test circuit, the system circuit is simplified, crosstalk between array units is reduced, and the test performance of the sensor system is improved, achieving rapid and high-precision measurement of the eddy current sensor array. Eddy current sensors are non-destructive, non-contact sensors based on the eddy current effect. Due to their excellent testing performance, they are widely used in mechanical quantity measurement and non-destructive testing of metallic materials. Research on eddy current sensor array testing technology began in the mid-1980s, and a number of documents and patents on eddy current array testing emerged in the late 1980s and early 1990s. Over the past decade, with the development of sensor technology and the improvement of processing technology, the research and application of eddy current sensor array testing have greatly advanced. They are not only used to measure the displacement of large-area metal surfaces, but also, due to their advantage of simultaneously detecting defects in multiple directions, are widely used in the inspection of metal welds, fatigue, aging, and corrosion detection of aircraft metal components, and non-destructive testing of turbines, steam generators, heat exchangers, and pressure vessel pipelines. Using array sensors, high-speed measurements over large areas can be achieved without the need for mechanical probe scanning, while maintaining the same measurement accuracy and resolution as a single sensor. This effectively improves the testing speed, measurement accuracy, and reliability of sensor systems. Furthermore, the flexible and diverse structural forms of sensor arrays allow for convenient inspection of parts with complex surface shapes. Therefore, the research of array sensors has become an important content and development direction in current sensor technology research. This paper studies the array testing technology of eddy current sensors based on a flat, flexible eddy current sensor array, employing a time-division multiplexing array testing method to achieve proximity measurement of large-area metal surfaces. By rationally designing the coil array and lead wire structure of the sensor probe, and cooperating with subsequent processing circuits and computer control, the rapid and high-precision measurement of the eddy current displacement sensor array and the real-time monitoring of the position status of large-area metal curved surface components are achieved. 1. Basic Principle of Eddy Current Detection The working principle of eddy current detection is to detect the interaction between the magnetic field of the excitation coil and the magnetic field of the induced eddy current. When an alternating current is passed through the sensitive coil, an alternating magnetic field is generated around the coil, as shown in Figure 1(a). If a metal conductor is moved into this alternating magnetic field, eddy currents will be induced on the surface of the conductor. These eddy currents will generate a magnetic field, the direction of which is exactly opposite to the direction of the original coil magnetic field, thus weakening the original magnetic field. [align=center][img=500,172]http://www.e-works.net.cn/images/128007859267656250.GIF[/img] Figure 1 Principle of Eddy Current Detection[/align] Eddy current sensors typically have two detection methods. One method is single-coil detection, which reflects changes in the magnetic field by detecting changes in the impedance of the sensitive coil. The equivalent impedance z of the coil can generally be expressed as a function: Z = F(σ, μ, f, x, r), where σ and μ are the conductivity and permeability of the measured metal conductor, respectively; f is the frequency of the excitation signal; x is the distance between the coil and the metal conductor; and r is the size factor of the coil, which is related to the coil's structure, shape, and size. It is evident that the change in coil impedance completely and uniquely reflects the eddy current effect of the measured metal conductor. In actual testing, controlling unwanted influencing factors allows for the detection of a specific quantity in the above formula. As a proximity sensor, the distance between the coil and the metal rake is directly related to the coil's impedance. When detecting defects on or near the surface of the metal, the presence of these defects will cause changes in the conductivity and permeability of the measured conductor, thus altering the coil's impedance parameters. Another method is dual-coil detection, as shown in Figure 1(b), which uses another coil as the detection coil to detect the superposition effect of the two magnetic fields. According to Faraday's law of electromagnetic induction, an induced electromotive force will be generated in the detection coil: [img=101,36]http://www.e-works.net.cn/images/128007859771406250.gif[/img] Where: ψ is the magnetic flux of the alternating magnetic field through the coil; n is the number of coils. The change in the magnetic field can be easily obtained by measuring the voltage generated in the detection coil. 2. Eddy Current Array Testing Technology 2.1 Eddy Current Array Form Compared with some other sensors, eddy current sensors have a prominent advantage—the probe structure is very simple. As can be seen from the basic principle of eddy current detection, the key component of the eddy current sensor probe is the sensitive coil. Therefore, eddy current array testing generally uses the coil array method, rather than arranging multiple independent sensor probes into an array. For different test conditions and technical specifications, the coil array can be designed with different structures and forms to realize the detection of complex surface parts, but the targeted design of the coil array and its matching circuit also brings relatively expensive costs. Although the design of eddy current coil array structures is flexible and varied, they can still be broadly classified into two typical array types based on their detection methods. ① Eddy current arrays based on single-coil detection, as shown in Figure 2(a), generally involve fabricating multiple sensitive coils directly on the substrate material and arranging them in a matrix array. Sufficient space must be maintained between adjacent coils to eliminate interference. This type of eddy current array is mostly used for proximity measurements on large-area metal surfaces, detecting the position of components, surface morphology, coating thickness, and the inner and outer diameters of rotating parts. It can also be used to detect surface defects such as cracks. ② Eddy current array detection based on a dual-coil method, generally designed as a large excitation coil plus a large array of small detection coils, as shown in Figure 2(b). It can very effectively detect defects in multiple directions on large-area metal surfaces, offering significant advantages in non-destructive testing applications and has largely replaced single-coil detection. In addition, a new type of magnetometer array L1 based on the eddy current effect has emerged in recent years. It is actually an array type based on dual-coil detection, achieving better testing performance through special design of the excitation coil and detection coil array structure. [align=center][img=490,175]http://www.e-works.net.cn/images/128007859976093750.GIF[/img] Figure 2 Eddy Current Array Form[/align] 2.2 Testing Methods for Eddy Current Arrays The fast response speed of eddy current sensors makes them well-suited for electronic scanning testing. By controlling an analog switch, all array units are scanned one by one, achieving the detection of all sensitive units in the sensor array. Using the scanning sampling method greatly simplifies the subsequent circuitry of the sensor, reduces system cost, and facilitates the miniaturization of the sensor system. However, the introduction of the analog switch also leads to a decrease in the sensor's testing accuracy. The key to testing eddy current sensor arrays lies in the lead design of the coil array. Figure 3 shows several commonly used lead design patterns based on the single-coil detection principle. Figure 3(a) shows a design pattern where the sensitive coil is led at both ends separately, and Figure 3(b) shows a row-column vertical scanning lead design. The purpose of adopting a row-column vertical structure is to reduce the number of external leads of the sensor array, which is of great significance for the practical application of sensors. However, it inevitably introduces crosstalk between array units, reducing measurement accuracy. The simplest way to solve the crosstalk problem is to lead each array unit separately (Figure 3(a)). Alternatively, the preprocessing circuit can be integrated with the sensitive unit to form an integrated sensor unit, but this comes at the cost of high manufacturing costs. Sometimes, to improve the scanning sampling speed, a row scanning sampling method is used, that is, scanning one row at a time, which greatly improves the testing speed of the sensor array, but requires increasing the number of preprocessing circuits. [align=center][img=500,164]http://www.e-works.net.cn/images/128007860256093750.GIF[/img] Figure 3 Eddy Current Array Test Method[/align][align=left] The eddy current sensor array in this paper is used to realize proximity measurement of metal surfaces, so a single-coil detection array is adopted. The test method is a compromise between the two design schemes mentioned above. One end of all coils is used as a common terminal and grounded, while the other end is led out and connected to the analog switch, as shown in Figure 3(c). Compared to vertical row and column leads, this design method, although having more leads, can greatly reduce crosstalk between coils and improve measurement accuracy. This advantage is particularly obvious for test systems with a small number of arrays. Moreover, using this lead method, the analog switch can use a one-to-many multiplexer, which can effectively simplify the control circuit and further miniaturize the subsequent processing circuit. 3. System Implementation and Experimentation 3.1 Design and Fabrication of Eddy Current Array Based on the above testing methods, a flat, flexible eddy current sensor array was designed and fabricated, as shown in Figure 4, which is a simplified structural diagram of the sensor probe. The sensor probe consists of a sensitive coil array and a lead cable formed by the coil leads. In general use, the coil array of the probe is often designed as a planar rectangular structure similar to Figure 2(a) (Figure 4(a)). Here, to more conveniently and reliably achieve the detection of complex shapes, the coil array is designed as a strip-shaped bifurcated structure (Figure 4(b)). Moreover, the shape of the coil array can vary depending on the shape of the surface being measured. In Figure 4(b), the coil array is designed as two parallel strip structures to achieve the detection of metal pipes. If the surface being measured is a general smooth surface, a vertical strip structure or a 6-branched structure can also be considered. The lead of each sensitive coil in the coil array is designed as a long, thin, flat lead cable, which is very important in applications involving small displacement measurements between large-area curved surfaces. The lead cable is directly connected to the processing circuit via a plug, and the cyclic scanning test of the sensor array is controlled by a computer. [align=center][img=443,179]http://www.e-works.net.cn/images/128007860556562500.GIF[/img] Figure 4 Design of the sensor array[/align] The sensor probe is fabricated on a polyimide film using flexible printed circuit board (FPCB) technology. The overall size of the probe's sensitive coil array is large, reaching 200 mm × 200 mm, while its thickness is very thin, not exceeding 0.15 mm, and it has good flexibility, making it applicable to the measurement of almost various geometric shapes. 3.2 Test System and Experiment The working principle of the test system is shown in Figure 5. A computer-controlled multiplexer cyclically scans and samples all array units. The sensor's conversion circuit uses a frequency-modulated oscillation circuit. The frequency signal output by the circuit is acquired by a frequency data acquisition card built into the computer. The acquired data is then sent to the computer for processing to obtain the desired position of the measured surface. [align=center][img=243,150]http://www.e-works.net.cn/images/128007860806562500.GIF[/img] Figure 5 Schematic diagram of the sensor system[/align] A large-area iron plane was used to calibrate the sensor array. The experimental results are shown in Figure 6. The four curves in Figure 6 represent the test results of four sensitive units on one branch of the sensor array. The reason why the four curves do not overlap is due to the influence of the sensitive coil position and the lead wire. This can be well solved by using software compensation. The test results show that within a measurement range of 2 mm, the test accuracy of the sensor array is better than ±0.25. When using the sensor array for testing, the center frequency of the oscillation circuit will decrease due to the influence of parasitic capacitance between array units, resulting in a slight decrease in the sensor's sensitivity. To improve the sensor's sensitivity, the oscillation circuit was optimized and improved, increasing its center frequency. Experimental results show that within a 2 mm measurement range, the sensor's average measurement sensitivity is approximately 70 Hz/μm, achieving excellent results. [align=center][img=258,192]http://www.e-works.net.cn/images/128007861096406250.GIF[/img] Figure 6 Performance experiment of the eddy current array[/align] 4. Conclusion Eddy current sensor arrays possess superior testing performance and broad application prospects compared to traditional eddy current sensors. This paper designs a flat, flexible eddy current sensor array to achieve real-time monitoring of the position of large-area curved metal components. Through research on eddy current array testing technology, a time-division multiplexing scanning detection method and signal transmission method were adopted. One end of all coils in the sensor coil array was designed as a common ground terminal, and the other end was connected to a multiplexer. Sampling of all sensitive coils in the sensor array was completed using only one set of signal conversion and signal conditioning circuits. This not only simplified the system circuit but also reduced interference between array units, improved the performance of the sensor system, and achieved rapid and high-precision measurement of the eddy current sensor array. Experimental results show that within a measurement range of 2 mm, the sensor array achieves a testing accuracy of ±0.25 and has high sensitivity, meeting the application requirements for real-time position monitoring of large-area curved metal components.