Control Modes and Gain Adjustment of Servo Motors

1. Torque Control

Torque control mode sets the output torque of the motor shaft externally through input of an analog quantity or direct assignment of an address value. Specifically, for example if 10V corresponds to 5Nm, when the external analog quantity is set to 5V the motor shaft outputs 2.5Nm. If the load on the motor shaft is less than 2.5Nm, the motor rotates forward. When the external load equals 2.5Nm, the motor does not rotate. When the load exceeds 2.5Nm, the motor rotates in reverse (this generally occurs under conditions with gravity loads). The set torque can be changed instantly by changing the analog quantity setting, or the corresponding address value can be changed via communication to achieve this. The main applications are for devices that require strict control of forces on materials, such as winding devices or fiber pulling equipment where the torque setting needs to be adjusted in real time according to changes in the winding radius to ensure the material is not subjected to changing forces as the winding radius changes.

2. Position Control

Position control mode generally determines the rotation speed by the input pulse frequency externally and determines the rotation angle by the number of pulses. Some servos can assign speed and displacement directly via communication.

Since position mode can provide very strict control over speed and position, it is generally used for positioning devices.

3. Speed Mode

Rotation speed control can be achieved by analog input or pulse frequency. Speed mode can also perform positioning with an external upper controller PID control in the outer loop, but the position signal of the motor or direct load must be fed back to the upper controller for calculation. Position mode also supports direct detection of the position signal of the final load, in which case the encoder on the motor shaft only detects motor speed, and the position signal is provided by the detecting device directly installed on the final load. The advantage is that it can reduce errors in the intermediate transmission process and increase the positioning accuracy of the entire system.

4. Full Closed-loop Control Mode

Full closed-loop control refers to closed-loop control relative to half closed-loop control.

First, let's understand half closed-loop control. Half closed-loop is a control system or PLC that issues speed pulse commands. The servo receives the command and executes it. During execution, the encoder on the servo provides position feedback to the servo itself for deviation correction. The servo itself can avoid errors, but mechanical errors cannot be avoided because the control system does not know the actual position. Full closed-loop control refers to the upper controller issuing velocity controllable pulse commands to the servo. The servo receives and executes the signal. During execution, a position feedback device is installed on the mechanical device to directly feedback to the control system. The control system compares and judges the deviation from the actual position to give the servo deviation correction commands, thus completing the speed loop control of the servo with the frequency controllable pulse signals from the controller and the position loop control of the servo with the position sensor (scale, encoder). This control mode that organically combines the servo motor, motion controller and position sensor is called full closed-loop control.

The Impact of PID Three-loop on Servo Control

Servo motors are generally controlled by three loops, that is, three closed-loop PID regulation systems. From the inside out, they are the current loop, speed loop and position loop.

1. Current loop: The innermost PID loop is the current loop. This loop is completely implemented in the servo driver using Hall elements to detect the output current of each phase in the driver and provide PID regulation based on negative feedback to the current setting to make the output current as close as possible to the set current. The current loop controls the motor torque. The computation in the driver is minimum and the dynamic response is fastest in torque mode.

2. Speed loop: PID regulation is performed based on negative feedback using the signal detected by the motor encoder. The PID output here directly serves as the "current loop setting". In other words, any mode must use the current loop, as the current loop is fundamental to control by achieving corresponding control of speed and position through current (torque) control.

3. Position loop: This is the outermost loop, which can be constructed between the driver and motor encoder or between the external controller and motor encoder or final load, depending on the actual situation. Since the position control loop output directly serves as the "speed loop setting", computation in position control mode involves all three loops, making the system computation the largest and dynamic response the slowest.

Adjustment of Loop Gains

1. First the current loop: The input to the current loop is the output from the PID regulation of the speed loop, which we call the "current loop setting". The difference between this "setting" and the "current loop feedback" value undergoes PID regulation in the current loop to output to the motor. The "current loop output" is the phase current of each motor phase. The "current loop feedback" comes from Hall elements (magnetic field induced to current voltage signals) installed on each phase rather than from the encoder feedback and is fed back to the current loop.

2. Speed loop: The input is the output from the PID regulation of the position loop plus the feedback of the position setting, which we call the "speed setting". The difference between this "speed setting" and the "speed loop feedback" value undergoes PID regulation (mainly proportional gain and integral processing) in the speed loop to output the above-mentioned "current loop setting". The speed loop feedback comes from the value obtained by the "speed arithmetic unit" after feedback from the encoder.

3. Position loop: The input is the external pulses (usually the case where data is written directly to the driver address is excluded for servos). The external pulses undergo smoothing filter processing and electronic gear calculation to serve as the "position loop setting". The setting and the pulse signal fed back from the encoder after counting calculation by the deviation counter undergo PID regulation (proportional gain adjustment, no integral or differential loops) in the position loop to output along with the position setting feedback to constitute the above-mentioned "speed loop setting". The position loop feedback also comes from the encoder installed on the trailing end of the servo motor and has no connection with the current loop - it samples from motor rotation rather than motor current, with no relation to the input, output or feedback of the current loop. The current loop is formed within the driver even without the motor, as long as analog loads (such as light bulbs) are installed on each phase.

The Concept of PID Control

PID is an important parameter in control systems, referring to the control method and the response mode between output and input - the three English letters standing for proportional (P), integral (I) and derivative (D).

PID control compares collected data with a reference value, then uses this difference to calculate a new input value. The goal of this new input value is to allow the system data to reach or maintain the reference value. Unlike other simple control calculations, PID control can adjust the input value according to the historical data and rate of appearance of differences, making the system more accurate and stable. It can be mathematically proven that a PID feedback loop can maintain system stability in situations where other control methods lead to steady-state errors or repeated processes in the system. In simple terms, PID control is feedback control - it measures the variable of interest, compares it to the setpoint, and uses the error to correct the control system.

The Impact of Individual PID Elements on the Difference

1. Proportional (P) alone proportionately computes the difference. Its prominent feature is it has correction for differences - meaning the adjusted quantity cannot be precisely equal to the set value after the adjustment process ends, there must be some residual, and the specific residual value can be calculated based on the proportional relationship. Increasing the proportion will effectively reduce the residual and increase system response, but is prone to cause vigorous oscillation in the system or even instability.

2. Integral (I) alone causes the change rate of the adjuster's output signal to be proportional to the difference signal. It is not difficult to understand that if the difference is large, the change rate of the integral part will be large. The proportional constant of this part, which we usually call the integral time constant in servo systems, means the smaller the integral time constant, the faster the system change rate. Therefore, increasing the integral rate (shortening the integral time constant) will decrease the stability of the control system until divergence oscillation occurs at last. The biggest advantage is that the adjusted quantity will have no residual in the end.

3. PI combines the advantages of P and I, utilizing P to rapidly negate interference impacts while using I to eliminate residuals.

4. Derivative (D) alone adjusts according to the direction and size of the rate of change of the difference. The adjuster's output is proportional to the differential of the difference over time. The derivative part can only play an auxiliary adjustment role and is often combined with other regulations to form PD and PID regulations. Its advantage is that it can adjust according to the change rate of the adjusted quantity (difference) without waiting for a large deviation to occur, in a sense giving the adjuster a degree of foresight to increase the system's response to small changes.

The PID constants of the servo current loop are generally preset optimally in the driver by the manufacturer, and do not need to be changed by the operator.

The speed loop mainly performs PI (proportional and integral) adjustment. The proportional part refers to the gain. Therefore, we need to properly adjust the speed gain and speed integral time constant to achieve the ideal effect.

The position loop mainly performs P (proportional) adjustment. For this, we only need to set the proportional gain of the position loop.

The adjustment of parameters for the position loop and speed loop do not have fixed numerical values. They need to be determined based on many condition factors, including the mechanical transmission connection method of the external load, movement mode of the load, inertia of the load, speed and acceleration requirements, rotor inertia and output inertia of the motor itself. A simple adjustment method is to gradually increase the gain parameters and gradually decrease the integral time constant within the empirical range based on the external load condition, with the stable state value without vibration overshoot being the optimal value.

When position mode needs to adjust the position loop, it is best to first adjust the speed loop (with the proportional gain of the position loop set to its minimum empirical value). Adjust the speed loop for stability first before adjusting the position loop gain gradually in small amounts. The response of the position loop is better slightly slower than the speed loop to avoid speed oscillations.

Principles and Precautions for Gain Adjustment

However, some servos require manual adjustment. When manually adjusting, note the following points: the position loop adjusts static gain, while the speed loop adjusts dynamic gain.

Simply put, adjust the position loop when the motor is stopped, and adjust the speed loop when the motor is running.

The position loop gain improves the speed of position response, i.e. finding the speed of reaching the position. The higher the gain, the shorter the time to reach the target, which is not related to speed. The closed-loop system has a high-speed oscillation process in the final location. It oscillates rapidly near the target value and finally finds the target. A higher gain means a faster end to this oscillation, which is one of the important performance indicators of servo motors.

The speed loop gain, of course, corresponds to speed and the performance of reaching the target speed.

The gain appears to be higher the better, but in actual operation this is not the case. Too high a gain for the servo system will cause resonance, producing huge noise and causing the motor to vibrate violently. It is recommended to adjust the gain as low as possible. The motor will not malfunction. Because most users of servos do not require very high responsiveness - only ensuring the motor does not resonate is needed.

Too high a gain will also cause problems such as overspeed, overload and overcurrent, because the ideal calculated values still differ from the actual capabilities of the motor, including the current loading and response capabilities of the electronic components.

Conclusion

The precise control of servo motors through various modes and optimized PID tuning opens up a world of possibilities for automation. Mastering these concepts allows engineers to solve complex mechanical problems by integrating intelligent motion control seamlessly.

Whether minimizing vibrations for delicate material handling or synchronizing multi-axis robotic movements with micrometer precision, servo technology transforms what was once done manually into integrated automated solutions. industries continue progressing.

More than just motors that spin or position components, servos are the invisible hands that make dreams of advanced manufacturing a reality. Through applications like rapid printing of parts on 3D printers or machining of intricate molds, they expand what can be created.

As loads increase in size and complexity, the ability to fine-tune three interlinked loops within servos maintains smooth operation. Proper tuning avoids damage while maximizing speeds and throughputs. This continuous improvement drives progress.

More powerful and efficient than simpler controls, PID-optimized servos are building blocks of automation that improve lives. They load medicines, assemble electronics and transport goods with minimal human involvement.

With further understanding and application of control modes coupling electric, electronic and mechanical systems, new vistas of industry automation and beyond will unfold. May these intelligent machines continue to amaze us with what they can achieve through motion.