I. Description of the phenomenon
The mechanical transmission part of an AC servo system inherently possesses a certain degree of elasticity, and coupled with a slight hysteresis in the position loop, it is prone to causing jitter at the mechanical end, affecting the tracking performance of the servo system and the positioning accuracy of the machine. This jitter phenomenon was discovered during research on the application of the Leadshine L5Z servo system in a robotic arm, as shown in Figure 1: after the positioning command is completed, even when the motor itself is nearly stationary, the mechanical transmission end still exhibits continuous oscillation.
Figure 1. Research platform for robotic arm applications
II. Application Problem Analysis
The figure below shows the oscillation phenomenon captured by the oscilloscope when acquiring signals such as position, speed, and current. It can be seen that after the command is completed, there are large fluctuations in each acquired signal, but the amplitude gradually decreases under the damping effect of the system.
Figure 2. Servo system output waveform before suppression
By capturing data and analyzing the spectrum of the telescopic shaft at the end of the command, it was found that the position, velocity, and current all contain low-frequency components of the same frequency, which is approximately around 10Hz.
Figure 3. Servo system output waveform after suppression
The Leadshine L5Z servo system's vibration suppression function addresses mechanical end effector swaying or low-frequency resonance in machine tools by suppressing the frequency of motion vibrations. Two vibration reduction frequencies are available, which can be used simultaneously, ranging from 1-200Hz. For detailed specifications, please refer to the Leadshine L5Z product manuals Pr2.14 and Pr2.16, as follows:
The specific steps are as follows:
Figure 4. Steps for using the L5Z vibration suppression technology.
By comparing the positioning time curves before and after vibration suppression, the change in the settling time can be calculated (based on a certain PUU as the positioning accuracy). The comparison results are shown in the table below. It can be seen that without vibration suppression, the settling time exceeds 314ms, while with vibration suppression, the settling time is reduced to 150ms.
Table 1 Comparison of settling time before and after vibration reduction
This servo system uses FFT to detect low-frequency vibrations in position, velocity, and current. After obtaining the vibration frequency information, it uses a low-frequency suppression filter to obtain a suppression compensation value, and adds this compensation value to the damping parameters for vibration suppression.
Figure 5. Schematic diagram of the oscillation suppression technology system
III. Principle Exploration and Experimental Analysis
The flexible connection device in a servo system can be approximated as a dual-inertia mechanical transmission system, as shown in the figure below.
Figure 6. Simplified schematic diagram of a dual-inertia mechanical transmission system
The differential equations of a typical two-inertia mechanical transmission system are as follows:
By establishing the differential equation of the system and the transfer function relationship between the motor/load speed and the electromagnetic torque, under certain assumptions, the root locus diagram of the servo system can be used to compare the end-position jitter frequency before and after suppression. The root locus before and after suppression is shown below. Reference 1 points out that the essence of suppressing the end-position jitter is to eliminate the lightly damped oscillation at the B pole.
Figure 7. Root locus diagram of the servo system before suppression
Figure 8. Root trajectory diagram of the servo system after vibration suppression
As a high-precision external laser sensor, the laser vibrometer is of great reference value for evaluating vibration characteristics. This experiment uses the laser vibrometer to collect vibration data from test points such as the mechanical end and the motor shaft end for vibration analysis. The experimental test curves are shown in the figure below:
Figure 9 Comparison curves of vibration suppression effect at the mechanical end.
Figure 10 Vibration spectrum analysis of the mechanical end.
Comparing the vibration data of the mechanical end, it can be seen that before suppression, the maximum peak value of the velocity fluctuation is 52 mm/s, and the main vibration frequency is 10.11 Hz. Under the action of servo damping, the amplitude gradually decreases. After suppression, the maximum value of the peak value of the velocity fluctuation is reduced to 7 mm/s, but it still contains a small amount of relatively high-frequency vibration.
Figure 11 Comparison curves of motor shaft end vibration suppression effect
Figure 12 Vibration spectrum analysis of motor shaft end
Comparing the vibration data at the motor shaft end, it can be seen that the speed fluctuation decreased from 7 mm/s to 1.4 mm/s after suppression, but the vibration frequency is similar to that of the mechanical end, which is 10.69 Hz. Analysis shows that the vibration of the mechanical end is forced vibration, and the vibration source comes from the vibration of the motor shaft end. Through the telescopic shaft, the vibration amplitude of the mechanical end is increased.
IV. Conclusion
After a robotic arm completes a positioning command, its end effector is prone to jitter, affecting positioning accuracy. However, by employing the Leadshine L5Z's automatic low-frequency suppression function, the impact of end effector vibration on positioning accuracy can be effectively reduced. Furthermore, after using the suppression function, the vibration amplitude is significantly reduced, and the settling time is also greatly shortened, improving the servo's response performance. Experimental data analysis has provided excellent verification of this function. In addition, the vibration suppression function of the Leadshine L5Z servo system has also been successfully applied to other industries such as Scara robots and electronic equipment.
About the author:
Cao Tong
Doctoral degree in Precision Instruments and Machinery from Harbin Engineering University
Intermediate Engineer
Director of Key Supported Laboratory in Shenzhen
Established the first applied research laboratory in China's motion control industry
Published numerous papers and patents in major domestic and international journals.
Main references:
1. Suppression of jitter at the positioning end of a permanent magnet AC servo system, 2015, *Journal of Electrical Machines and Control*
2. Vibration suppression intwo-mass drive system using PI speed controller with different additionalfeedbacks.2006
