Principle of open-loop control of stepper motor

Principle of open-loop control of stepper motor

When a DC current flows through one phase of the stator winding of the stepper motor, the rotor tooth closest to the phase is attracted by the stator phase, and the electromagnetic torque generated is greater than the load torque, causing the rotor to move. When the rotor rotates to the position where the electromagnetic torque and the load torque are balanced, the rotor stops moving, and this electromagnetic torque also turns the load to the position where it needs to be positioned. Then the excitation current is applied to the next phase, and the other rotor tooth closest to the phase is attracted. The load is driven by the electromagnetic torque of this phase and moves 1 step angle to reach the next static position. The number and frequency of switching of the excitation phase determine the final angle and speed of the rotor rotation. The product of the number of switching phases and the step angle is the step angle (the technical term is the angle added by the step action), which determines the final static position. Relative to the load torque, if the torque generated by the stepper motor is large enough, the switching command can drive the load for position control. At this time, the position balance force is generated by the static torque of the stepper motor.

The following figure shows the torque-angle characteristic curves of each phase of a two-phase PM stepper motor. When the "bar A" phase winding is excited, the loaded rotor must be displaced, and the load should be within the force range between the rotor and phase A. The positional relationship between the stator and the rotor when the "bar A" phase excitation winding is energized is shown in the upper part of the figure. The torque-angle characteristic of the excitation phase "bar A" is represented by a solid curve; when the other phase windings are excited, the torque-angle characteristic curves generated are represented by dotted lines.

When lightly loaded or unloaded, the static torque is determined by the position, so the "bar A" phase torque moves along the direction of the curve arrow to the intersection point C1 with the horizontal axis; in fact, the rotor stops at the load balance point on the torque curve.

In turn, if phase B is energized, the rotor stops at point b1, and the angle difference between b1 and C1 is the step angle.

The speed control can use the open loop control (OPENLOOP) method. Changing the speed only requires changing the switching frequency command, which is equivalent to the function of a variable frequency synchronous motor.

Key points of open-loop control of stepper motors

1. Current loop

The motors on the market currently generally have small inductance and resistance. They are rarely driven by voltage like old steppers. They are mainly controlled by controlling the AB phase current. The current is a sine wave with a phase difference of 90 degrees. The drive mostly adopts the H-bridge control method, and two H-bridges control the two-phase current. There are two ways to control the current. One is AC control, and the AB phase current is independently controlled. One gives a sine reference signal and one phase gives a cosine reference signal. The other is DC control, which is similar to vector control. The coordinates of the AB phases are transformed to control the current of DQ. As far as I know, most of the drives on the market are mainly AC control.

2. Speed ​​identification

The medium speed of stepper motor is prone to oscillation, and damping compensation is often performed through speed or acceleration identification to improve the stability of medium and high speeds. There are three types of speed identification that I have tested, namely sliding film control, PI phase-locked loop, and no PI phase-locked loop. The latter two have the best effect. The speed difference can suppress oscillation by compensating the q-axis current or the lead angle, and the effect is theoretically the same. The method of speed identification can refer to the sensorless control of pmsm. TI, ST, and microchip all have similar solutions.

3. Back EMF compensation

The number of pole pairs of a stepper motor is usually 50, so the back-EMF coefficient is relatively large. Usually above 500rpm, the motor is already working in the stage of weak magnetic field speed increase, so proper back-EMF compensation will help improve the response speed of the motor. In speed identification, back-EMF data is usually required for speed and position identification. Back-EMF compensation can use the conclusions and intermediate quantities of speed identification.

4. Parameter identification

There are a huge number of stepper motors on the market, and different manufacturers have different parameters. The versatility of the driver requires that the driver can drive different motors. The speed identification and the PI parameter setting stage of the current loop generally require resistance and inductance data. The method I use is to give a constant current when power is turned on, turn off all tubes to let the current flow naturally, and test the time it takes for the constant current to drop by 1A to calculate the time constant. By giving a second constant current, the voltage difference between the two constant currents can be used to find the resistance without dead zone influence, and then calculate the inductance value.

5. MCU selection

The choice of MCU mainly focuses on the computing power. I have tested it on TMS320F28027, 034, STM32F103, and xmc1302 platforms.