Abstract: This paper introduces a stepper motor testing system based on virtual instrument technology. It utilizes virtual instruments to replace traditional instruments and testing software to replace hardware circuits, enabling multi-channel data acquisition, analysis, and storage, thus achieving online testing and fault diagnosis of the stepper motor system. Keywords: Virtual instrument; Stepper motor; Data acquisition 1 Hardware Composition The testing system mainly consists of a signal interface and a virtual instrument (Figure 1). The signal to be tested is led out from the interface on the control cabinet, selected and conditioned, and then sent to the industrial computer. The data acquisition card acquires the data, and the data processing software analyzes, displays, and stores it. The stepper motor system consists of a pulse controller, a drive circuit, and the stepper motor itself. Depending on the testing requirements, such as routine testing, real-time monitoring, and fault diagnosis, the output of the pulse controller, the output of the drive circuit, and the signals of the motor windings need to be tested separately. To more effectively utilize the hardware resources of the acquisition card and the data processing capabilities of the computer, a signal selection circuit is set in the interface section to send the signals to be tested to the subsequent system. The interface circuit structure is shown in Figure 2. Different combinations of two selection switches allow for the extraction of pulse controller signals from the driver board input stage, drive circuit signals from the driver board output stage, and stepper motor winding current signals from the motor circuit, respectively. The signal conditioning circuit uses an operational amplifier to perform differential operations on the signals across the sampling resistors, obtaining voltage and current signals, which are then output to the data acquisition card in single-ended mode. Stepper motors are often driven by square wave voltages, which contain certain high-frequency components and are large-amplitude signals with abrupt changes. Therefore, the LM318 high-speed broadband operational amplifier is selected, with a gain bandwidth of 15MHz and a slew rate of 70V/μs. To further improve the signal-to-noise ratio of the measured signal, reduce the difficulty of software data processing, and reduce computational load, two 1000μF electrolytic capacitors are added to the power supply section of the LM318 for decoupling, and a 0.2μF ceramic capacitor is added to its output to filter out high-frequency noise. The hardware of the virtual instrument adopts the DAQ data acquisition system based on PCI bus technology. The selected PCI-6071E data acquisition card can realize multi-channel parallel detection of 32 stepper motors and their drive circuits and pulse controllers. 2. Software Design Based on the modular programming concept, the detection program (Figure 3) is structured from top to bottom into a main program layer, a logic layer, and a driver layer. The main program layer consists of the user interface and the test execution part. The logic layer is responsible for verifying logical relationships and making relevant decisions. The driver layer is responsible for communication with the instrument, the device under test, and other applications. The software development platform is NI's LabVIEW. The main task of the detection program is multi-channel data acquisition, analysis, and storage. Therefore, program optimization and running efficiency are quite important. The software development utilizes several advanced programming technologies supported by LabVIEW, such as data flow, multithreading, timed loops, and state machines. 3. Signal Processing The essence of the virtual instrument is the digital processing of analog signals, specifically divided into online processing and post-processing. Online data processing mainly includes time-domain analysis of current, voltage, and pulses, which involves relatively small computational loads. For the system's operating status, the number of motor steps, the number of voltage cycles provided by the drive board, and the number of pulses emitted by the pulse controller are obtained by counting the corresponding signals. For the motor's operating parameters, the motor speed curve is obtained by measuring the current frequency, the acceleration curve is obtained by differentiating this, and the power curve is obtained by numerically integrating the current. On the other hand, a more detailed time-domain analysis is performed on the current signal to provide time-domain characteristic values ​​for system analysis. Peak Detector is used to detect the peak of the signal to obtain the value and position of the highest point in each cycle, and the characteristic curve of the motor is constructed based on this. When the motor is operating normally, the characteristic curve is approximately a horizontal straight line, while during abnormal operation, it will produce translation and fluctuations, and its mean and variance will change significantly. Pulse Parameters is used to detect the signal parameters to obtain parameters such as overshoot and rise time, which describe the detailed information of the current waveform. Therefore, the ratio of overshoot to amplitude, the ratio of rise time to frequency, and the mean and variance of the characteristic curve are selected as three sets of time-domain characteristic values ​​for system state analysis. Post-processing mainly includes frequency domain analysis of current and voltage. For stepper motor system testing, a crucial application is identifying and displaying transient anomalies embedded in normal signals. To overcome the lack of time resolution in Fourier transform, a short-time Fourier transform (SFT) analysis method is employed for abnormal signal segments. The signal algorithm performs SFT analysis to obtain the signal amplitude spectrum, indicating the energy distribution of winding current and drive voltage over a short time period. When an abnormality occurs in the motor system, the current and voltage signals, in addition to the normal fundamental and harmonic components, exhibit additional low-frequency components or DC components, and the amplitude ratio of the harmonic to the fundamental frequency also changes significantly. Therefore, the amplitude ratio of the third harmonic to the fundamental frequency is selected as the frequency domain characteristic value for system state analysis. For online detection and fault diagnosis systems, besides selecting appropriate signal processing algorithms to extract effective characteristic values, a more important aspect is the summarization and classification of historical data of the tested system, providing typical values ​​for each characteristic as criteria for system state judgment. The following are typical values ​​measured on the 1020 miniature two-phase permanent magnet stepper motor of ARSAPE in Switzerland. The driving method is two-phase single four-step drive using A3966SLB drive module. The stepper motor detection system developed based on virtual instrument technology improves the maintenance efficiency of the system and greatly shortens the fault recovery time by diagnosing the faulty unit when a fault occurs. References [1] Gary W. Johnson, Richard Jennings. LabVIEW Graphical Programming [M]. Beijing: Peking University Press, 2002. [2] Ma Jianming, Zhou Changcheng. Data Acquisition and Processing Technology [M]. Xi'an: Xi'an Jiaotong University Press, 2001. [3] Lü Yong, et al. Virtual Instrument Technology and Its Application in Mechanical Fault Diagnosis [J]. Journal of Wuhan University of Science and Technology, 2002, (2). [4] Wang Zongpei, Ren Lei, Shi Jingzhuo. Analysis of Winding Current Waveform of Five-Phase Hybrid Stepper Motor [J]. Micro-Electro-Mechatronics, 1997, (5).