A test device and a measurement method based on mechanical friction torque of a motor

Abstract

The present application aims to provide a kind of test equipment and measurement method based on motor mechanical friction torque.The present application includes industrial computer and motor, the industrial computer is connected with the motor via driver, the motor is connected with the product to be measured transmission via torque sensor, the torque sensor is connected with signal adapter plate, the signal adapter plate includes ADC acquisition board, the ADC acquisition board is connected with core control board, the core control board reads the angle voltage data and torque data collected by the ADC acquisition board, and the core control board is connected with the industrial computer in communication.The present application is applied to the technical field of test equipment.

Description

Technical Field

[0001] This invention relates to the technical field of testing equipment, and in particular to a testing device and measurement method based on the mechanical friction torque of an electric motor. Background Technology

[0002] With the rapid development of industrial automation, new energy vehicles, intelligent manufacturing, and robotics, various rotating electric motors have become core actuators in modern electromechanical transmission systems. From micro servo motors to large industrial drive motors, the mechanical performance of the motor directly determines the operating efficiency, dynamic response, energy consumption level, and service life of the entire system. In the design and manufacturing process of motor products, the mechanical friction characteristics between the stator and rotor are key indicators for evaluating assembly quality, lubrication status, bearing performance, and overall process consistency. The mechanical friction torque of a motor in the unenergized state typically includes bearing rolling friction, sliding surface friction, seal friction, air viscous resistance, and residual stress resistance torque caused by assembly interference. This is the fundamental resistance that the motor must overcome during no-load operation and is also an important parameter for measuring the motor's mechanical smoothness and energy efficiency.

[0003] The national standard GB / T 30549-2014, "General Technical Conditions for AC Servo Motors," clearly defines the test for static friction torque of motors. It requires that, with the motor windings open, a torque be applied to the shaft using a pulley and weight method or other equivalent method, and the minimum resistance torque required to start the motor rotor rotating must be measured at several equally spaced points. However, this standard and existing test methods are mainly designed for motors with permanent magnets. For electromagnetic motors that rely entirely on coil winding excitation and are not electrically excited during the test, high-precision testing of their mechanical friction characteristics still faces many technical challenges.

[0004] In permanent magnet motors, even when the motor is not energized, the magnetic interaction between the permanent magnets and the stator core generates a periodic torque, known as cogging torque. According to the national standard GB / T 30549-2014, cogging torque is defined as "the periodic torque generated during one revolution of the motor when the windings are open, caused by the slotting of the armature core, tending towards the position of minimum magnetic reluctance." Cogging torque is generated by the interaction between the magnetic field generated by the permanent magnets and the change in magnetic reluctance of the armature core slots; it is an electromagnetic effect rather than a mechanical friction effect. Therefore, when the motor is tested without permanent magnets and in an un-energized state, cogging torque does not exist, and the resistance measured by the torque sensor comes entirely from pure mechanical friction.

[0005] Existing methods for testing the mechanical friction torque of electric motors can be categorized as follows: 1. The static weight / pulley method is one of the recommended methods for testing cogging torque in the national standard GB / T 30549-2014, and it is also widely used for measuring the static friction torque of motors. Its basic principle is as follows: the motor is fixed on a test bench, a lever arm of a certain length is installed on the motor shaft, and a torque is applied to one end of the lever arm by suspending weights. The weight of the weights is gradually increased, and the mass of the weights at the instant the motor rotor begins to rotate is recorded. The torque value is then calculated using the torque balance formula. During the test, measurements are usually taken at multiple equally spaced points in both directions, and the maximum value is taken as the static friction torque result. This method is simple in principle, low in cost, and does not require complex testing equipment and sensors, making it mainly suitable for applications where high precision is not required. However, this method has obvious limitations: First, the measurement results include both friction torque and cogging torque, making it impossible to effectively separate these two different torque components; second, the use of discrete point measurement makes it impossible to obtain the torque change curve during continuous rotation, making it difficult to capture periodic fluctuations caused by stator slotting or bearing defects; third, the test operation relies on manual observation and operation, resulting in low testing efficiency, poor repeatability, and a long time required to complete the full range test for a single motor, with the accuracy of the results greatly affected by the operator's experience.

[0006] 2. The torque sensor dynamic measurement method is a more accurate method for cogging torque testing, and it is also defined in detail in the national standard GB / T30549-2014. Its basic structure is as follows: the prime mover reduces its speed to an ultra-low speed of 1~10 r / min through a reduction system, and drives the rotor of the motor under test to rotate through a coupling; the torque sensor is connected in series in the transmission chain to detect the torque signal on the transmission shaft in real time; the stator of the motor under test is fixed on the base, and the rotor is in a free-rotating state. The core idea of ​​the torque sensor dynamic measurement method is to use the high sampling rate of the torque sensor to continuously collect torque values, obtain a torque-time curve, and then obtain a torque-angle curve through angle conversion, thereby extracting the peak-to-peak value of the cogging torque. This method can obtain a relatively complete torque waveform and has high measurement accuracy. However, this method also has limitations: the test system structure is relatively complex, and the torque fluctuations of the prime mover and transmission system themselves will couple into the measurement results, requiring filtering to obtain an effective signal; more importantly, this method is mainly designed for permanent magnet motors, and the test results contain a mixture of cogging torque and friction torque signals, making it difficult to effectively separate the two. Furthermore, in the field of academic research, Yang Zhijian et al. from South China University of Technology published a paper entitled "A Method for Measuring Cogging Torque of Permanent Magnet Motors" in the 5th issue of the *Journal of Electrical Machines and Control* in 2022. This paper proposes a simple method for measuring cogging torque based on weight-driven operation and encoder acquisition of rotor position. This method uses weights to drive a disk mounted on the motor shaft to rotate the motor. An encoder acquires the rotor position signal, and the cogging torque is calculated according to the dynamic balance equation of the test system. Low-pass order filtering and angular Fourier transform are then performed in the angular domain to obtain the angular domain curve of the cogging torque, its order components, and amplitude. This method does not rely on a complex torque test bench and has the advantages of simple operation and low cost. However, it requires manual addition of weights during the measurement process, making it difficult to achieve continuous uniform speed drive, thus falling into the category of discrete measurement. Furthermore, the cogging torque and friction torque in the test results are not effectively separated.

[0007] 3. Alternatively, online parameter estimation can be used. This involves measuring the motor's operating parameters to estimate the friction torque. The motor is fixed to a motor performance testing system, and by measuring parameters such as input power, speed, and torque, combined with the motor's efficiency curve, the output torque is calculated, and thus the friction torque is estimated. This method does not require removing the motor from the transmission system and allows for real-time monitoring of the motor's friction torque. However, it is essentially an indirect estimation, and its accuracy depends on the accuracy of the motor model. Furthermore, the motor is in an energized state, and the measurement results include the influence of electromagnetic effects.

[0008] 4. In the field of precision shaft friction torque measurement, Feng Bing's master's thesis at Harbin Institute of Technology, "Research on Testing Methods of Friction Torque," systematically studied the measurement methods of friction torque, proposed a friction torque testing method based on virtual instruments, and built a torque testing experimental system based on torque sensors, photoelectric encoders, and low-speed AC motors. This research mainly focuses on the measurement of friction torque in purely mechanical shaft systems, testing shaft components rather than the complete motor stator-rotor assembly, and still requires independent encoders for angle measurement. In the field of mechanical transmission, some researchers have developed shaft friction torque testing benches using stepper motor drives and precision force sensors. By adjusting the installation position of the force sensor, the measurement range of the testing bench can be adjusted to test shaft components with different measurement ranges. However, these studies all focus on the shaft components themselves, rather than the comprehensive mechanical characteristics of the motor stator-rotor assembly.

[0009] Based on the above analysis of existing technologies, the following common technical shortcomings can be summarized: First, cogging torque and mechanical friction torque are not effectively separated. In permanent magnet motors, cogging torque and friction torque coexist in the measurement signal, and existing methods cannot effectively separate them. For electromagnetic motors without permanent magnets, cogging torque does not exist, but there is little systematic research in existing literature on high-precision measurement methods for pure mechanical friction torque of such motors under unexcited conditions. Second, angle information acquisition relies on external encoders. Whether measuring cogging torque or shaft friction torque, establishing a precise correspondence between torque and rotor angle requires additional angle encoders or photoelectric encoders, increasing the hardware cost and mechanical installation complexity of the test system. Third, there is a lack of dynamic harmonic analysis capabilities under continuous rotation. Discrete measurement methods such as the static weight method can only obtain torque values ​​at a limited number of discrete points, and cannot obtain the dynamic torque waveform and its frequency domain characteristics during a continuous rotation. Although the torque sensor dynamic measurement method can obtain continuous waveforms, it is sensitive to the torque fluctuations of the transmission system itself, and the standard torque meter configuration still requires an external angle encoder. Fourth, the testing efficiency is low. According to existing traditional testing methods, a motor needs to have its static friction torque measured at no less than 10 points. Even skilled engineers still need a long testing time, which is difficult to meet the cycle time requirements of modern motor production lines.

[0010] For example, Chinese patent CN121643540A discloses a motor control system and control method for a two-axis photoelectric pod. Establishing a precise correspondence between torque and rotor angle requires additional configuration of angle encoders or photoelectric encoders, increasing the hardware cost and mechanical installation complexity of the testing system. Therefore, it is necessary to provide a testing device and measurement method based on the mechanical friction torque of the motor. This device should be highly integrated, low-cost, and highly accurate, suitable for the quality inspection of semi-finished or finished products of electromagnetic motors without permanent magnets or coil excitation. It should also enable rapid location of defect sources through FFT harmonic analysis, thus improving the quality control level and process improvement efficiency in motor production. Summary of the Invention

[0011] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a test device and measurement method based on the mechanical friction torque of an electric motor. It has high integration, low cost and high accuracy. It is suitable for the quality inspection of semi-finished or finished products of electromagnetic motors without permanent magnets or coil excitation. It can quickly locate the source of defects through FFT harmonic analysis, which is conducive to improving the quality control level and process improvement efficiency of motor production.

[0012] The technical solution adopted in this invention is as follows: This invention includes an industrial control computer and a motor. The industrial control computer is connected to the motor via a driver. The motor is connected to the product under test via a torque sensor. The torque sensor is connected to a signal adapter board. The signal adapter board includes an ADC acquisition board. The ADC acquisition board is connected to a core control board. The core control board reads the angle voltage data and torque data acquired by the ADC acquisition board. The core control board is communicatively connected to the industrial control computer.

[0013] As can be seen from the above scheme, this application directly uses the A-phase, B-phase, and Z-phase TTL level pulse signals output by the incremental encoder integrated within the dynamic torque sensor. The A-phase pulses are accumulated and counted using an FPGA hardware counter. Combined with the known number of pulses per revolution (PPR, programmable setting, e.g., 8192), the rotor absolute angle is calculated in real time (resolution up to 360° / 8192≈0.0439°). Simultaneously, the time interval between adjacent A-phase pulses is measured using the FPGA's internal high-frequency clock (100 MHz), and the instantaneous rotational speed is calculated in real time. One Z-phase pulse is emitted per revolution, ensuring data alignment for each measurement.

[0014] In existing solutions, the torque signal is acquired by a torque sensor through an analog channel, while the angle signal is acquired by a separate external encoder through another channel. These two signals, originating from different sensors, undergo different signal conditioning paths and acquisition cards, resulting in uncertain millisecond-level time delays and phase shifts upon reaching the host computer. For applications requiring FFT frequency domain analysis, this leads to amplitude distortion and phase errors in harmonic components, severely impacting the accuracy of defect location. In this method, the torque analog signal and the TTL encoder signal originate from the same torque sensor and are mechanically coaxial. An FPGA is used as the core acquisition unit, and the current A-phase pulse counter value is synchronously latched in hardware at the instant the ADC conversion is completed (<4μs). The latching delay is <5 ns, and the overall torque-angle synchronization accuracy is better than 5 μs. At a rotational speed of 10 rpm, the corresponding angle synchronization error is less than 0.0003°, which is two orders of magnitude better than existing technologies (millisecond-level, error greater than 0.5°), providing high-quality paired data for high-precision order FFT analysis.

[0015] Existing conventional torque meter methods typically employ a multi-stage transmission chain consisting of a servo motor, a reducer (or harmonic reducer), a coupling, and a torque sensor. The frictional torque and periodic torque fluctuations (such as the cogging effect of the reducer) of the reducer, gears, and other transmission components are directly superimposed on the torque measurement signal. However, the frictional torque of the product under test is often very weak (in the mN·m range), and interference from the transmission system can completely drown out the true signal. Even with filtering, it is difficult to completely eliminate this interference at ultra-low speeds. This method uses a low-speed, high-torque servo motor for direct drive (without a reducer), and achieves stable operation at ultra-low speeds of 1-30 rpm through closed-loop speed control. The transmission chain contains only two flexible couplings, without any gears, synchronous belts, or reducers. Therefore, the additional frictional torque of the transmission system itself is extremely small (only the residual imbalance of the coupling itself), and there is no source of periodic torque fluctuations. Furthermore, the background noise spectrum can be obtained through no-load (without the product installed) testing and subtracted from the product test results to further eliminate residual interference. This allows this method to measure the true mechanical frictional torque between the motor's stator and rotor more purely.

[0016] This application offers significant advantages in system simplification, synchronization accuracy, transmission interference elimination, and testing efficiency. These advantages directly translate into user value: lower equipment costs, higher testing reliability, faster production line cycle time, and accurate attribution of motor quality problems, thereby improving the quality control level and process improvement efficiency of motor manufacturers.

[0017] In a preferred embodiment, the test device based on the mechanical friction torque of the motor further includes a first coupling, through which the motor is connected to the torque sensor.

[0018] In a preferred embodiment, the test equipment based on the mechanical friction torque of the motor further includes a second coupling, through which the torque sensor is connected to the product under test.

[0019] A preferred embodiment is that the measurement method includes the following steps: Step A: The industrial control computer sends a command to the driver to control the motor to rotate, which drives the torque sensor through the first coupling, thereby driving the product under test to rotate synchronously and coaxially, generating a torque force. Step B: The torque sensor outputs a torque analog voltage of ±5 VDC, corresponding to ±0.2 N·m, and simultaneously outputs A, B, and Z phase TTL pulse signals, with 8192 A phase pulses and one Z phase reference pulse per revolution; Step C: The above signal is connected to the signal adapter board through a shielded cable. The signal adapter board performs signal buffering, level conversion, distribution, and preliminary filtering. Step D: The signal adapter board processes the A / B / Z TTL pulses respectively and sends them to the digital input port of the ADC acquisition board; Step E: The ADC acquisition board converts the analog voltage into a digital quantity and sends the torque data to the core control board via the SPI bus; Step F: The core control board receives SPI data and TTL pulses from the ADC acquisition board. Phase A is used for counting and speed measurement, and phase Z is used for positioning. The SPI data is the torque value. Finally, the result is output to the industrial control computer to perform FFT and other calculations. FFT refers to Fast Fourier Transform. The final test result is generated and uploaded to MES, which stands for Manufacturing Execution System. Attached Figure Description

[0020] Figure 1 This is a hardware block diagram of the present invention; Figure 2 This is a waveform diagram of TTL level output and FPGA acquisition; Figure 3 This is a graph of load test data; Figure 4 This is a graph of no-load test data. Detailed Implementation

[0021] like Figure 1As shown, in this embodiment, the present invention includes an industrial computer 1 and a motor 2. The industrial computer 1 is connected to the motor 2 via a driver 3. The motor 2 is connected to the product under test 5 via a torque sensor 4. The torque sensor 4 is connected to a signal adapter board 6, which includes an ADC acquisition board 7. The ADC acquisition board 7 is connected to a core control board 8. The core control board 8 reads the angle voltage data and torque data acquired by the ADC acquisition board 7. The core control board 8 is communicatively connected to the industrial computer 1. The core control board 8 is a Xavier 7020.

[0022] like Figure 1 As shown, in this embodiment, the test device based on the mechanical friction torque of the motor further includes a first coupling 9, and the motor 2 is connected to the torque sensor 4 via the first coupling 9.

[0023] like Figure 1 As shown, in this embodiment, the test equipment based on the mechanical friction torque of the motor further includes a second coupling 10, and the torque sensor 4 is connected to the product under test 5 via the second coupling 10.

[0024] In this embodiment, the measurement method includes the following steps: Step A: The industrial computer 1 sends a command to the driver 3 to control the motor 2 to rotate, which drives the torque sensor 4 through the first coupling 9, thereby driving the product under test 5 to rotate synchronously and coaxially, generating a torque force. Step B: The torque sensor 4 outputs a torque analog voltage of ±5 VDC, corresponding to ±0.2 N·m, and simultaneously outputs A, B, and Z phase TTL pulse signals, with 8192 A-phase pulses and one Z-phase reference pulse per revolution. Figure 2 As shown; Step C: The above signal is connected to the signal adapter board 6 through a shielded cable. The signal adapter board 6 performs signal buffering, level conversion, distribution, and preliminary filtering. Step D: The signal adapter board 6 processes the A / B / Z TTL pulses respectively and sends them to the digital input port of the ADC acquisition board 7. Step E: The ADC acquisition board 7 converts the analog voltage into a digital quantity and sends the torque data to the core control board 8 via the SPI bus. The ADC acquisition board 7 includes a high-precision ADC (analog-to-digital converter, 16-bit, 1MSPS) and digital signal isolation. Step F: The core control board 8 receives SPI data and TTL pulses from the ADC acquisition board 7. Phase A is used for counting and speed measurement, and phase Z is used for positioning. The SPI data is the torque value. Finally, the result is output to the industrial computer 1 to perform FFT and other calculations. FFT refers to Fast Fourier Transform. The final test result is generated and uploaded to MES, which stands for Manufacturing Execution System.

[0025] like Figure 3 , Figure 4 The waveforms shown at the top are the torque data waveforms for one revolution, and the histogram in the lower right corner is the FFT analysis waveform of the torque data. The device uses an FPGA as its core to realize encoder pulse parsing, angle and speed calculation, ADC synchronous sampling, and Ethernet data upload with full time synchronization. All acquisition actions are based on the encoder mechanical angle, ensuring that the waveforms for each revolution and the data at each angle point are strictly corresponding and highly repeatable.

[0026] Although the embodiments of the present invention are described with reference to actual solutions, they do not constitute a limitation on the meaning of the present invention. Modifications to the embodiments and combinations with other solutions based on this specification will be obvious to those skilled in the art.

Claims

1. A testing device based on the mechanical friction torque of an electric motor, comprising an industrial control computer (1) and an electric motor (2), characterized in that: The industrial computer (1) is connected to the motor (2) via the driver (3). The motor (2) is connected to the product under test (5) via the torque sensor (4). The torque sensor (4) is connected to a signal adapter board (6). The signal adapter board (6) includes an ADC acquisition board (7). The ADC acquisition board (7) is connected to a core control board (8). The core control board (8) reads the angle voltage data and torque data acquired by the ADC acquisition board (7). The core control board (8) is communicatively connected to the industrial computer (1).

2. The testing equipment based on the mechanical friction torque of an electric motor according to claim 1, characterized in that, The test equipment based on the mechanical friction torque of the motor also includes a first coupling (9), and the motor (2) is connected to the torque sensor (4) via the first coupling (9).

3. The testing equipment based on the mechanical friction torque of an electric motor according to claim 1, characterized in that, The test equipment based on the mechanical friction torque of the motor also includes a second coupling (10), and the torque sensor (4) is connected to the product under test (5) via the second coupling (10).

4. A measurement method comprising the testing equipment based on the mechanical friction torque of a motor as described in claim 2, characterized in that, The measurement method includes the following steps: Step A: The industrial computer (1) sends a command to the driver (3) to control the motor (2) to rotate, and drives the torque sensor (4) through the first coupling (9), thereby driving the product under test (5) to rotate synchronously and coaxially, generating a torque force; Step B: The output torque analog voltage of the torque sensor (4) is ±5 VDC, corresponding to ±0.2 N·m. At the same time, it outputs A, B and Z phase TTL pulse signals, with 8192 A phase pulses and one Z phase reference pulse per revolution. Step C: The above signal is connected to the signal adapter board (6) through a shielded cable. The signal adapter board (6) performs signal buffering, level conversion, distribution and preliminary filtering. Step D: The signal adapter board (6) processes the A / B / Z TTL pulses respectively and sends them to the digital input port of the ADC acquisition board (7); Step E: The ADC acquisition board (7) converts the analog voltage into a digital quantity and sends the torque data to the core control board (8) via the SPI bus. Step F: The core control board (8) receives SPI data and TTL pulses from the ADC acquisition board (7). Phase A is used for counting and speed measurement, and phase Z is used for positioning. The SPI data is the torque value. Finally, the result is output to the industrial control computer (1) to perform FFT and other calculations. FFT refers to Fast Fourier Transform. The final test result is generated and uploaded to MES. MES is the Manufacturing Execution System.

Images