Engineering · Learn · Motion
Induction Motor Position Control
Can an ordinary AC motor stop at an exact position? Yes, with an encoder and a closed-loop VFD, within limits.
A plain VFD controls speed, not position. Add an encoder on the motor shaft and a drive with encoder feedback, and the drive can compare the commanded position with the real one and correct the difference. That gives you accurate indexing, rotary tables and long-travel axes using a standard squirrel-cage motor, at lower cost than a servo. This page shows how the loop works, what you need, how to wire it, what accuracy to expect and when a servo is the better answer.
Drive features differ by brand and model. Which position modes, feedback cards and parameters exist is set by your VFD manual, so treat this page as the idea and the manual as the rules. Pictures are illustrative drawings.
The controller (PLC, motion controller or HMI) sends a target position. The VFD drives the motor. The encoder on the shaft reports where the shaft really is. The drive compares the two and changes its output until the error is gone. Without the encoder, the drive only guesses.
Position control is a stack of loops, fastest inside. The innermost loop controls motor current (vector control). Around it is the speed loop, usually PI. Outside is the position loop, usually proportional, which turns position error into a speed command. Tune from the inside out: speed loop first, then the position loop.
Field-oriented control (FOC) treats the AC motor like a DC motor. Two measured phase currents are turned into a fixed two-axis frame (Clarke), then into a frame that rotates with the rotor flux (Park). There, one current component sets the magnetic flux (d) and the other sets the torque (q), and each gets its own PI controller. The result is turned back (inverse Park) and made into inverter switching pulses (SVPWM). The Park step needs the flux angle θ. In an induction motor it is the encoder’s rotor position plus the slip angle the drive calculates, which is why the drive needs the nameplate data and auto-tune.
Channel B leads or lags A by 90°, and the drive reads which one comes first to know the direction. The drive counts every edge of A and B, so counts per revolution = PPR × 4. Z gives one pulse per turn and is used to find a home position after power-up. An absolute encoder needs no homing, because it reports its position the moment it is powered.
| Encoder data | Typical range |
|---|
The encoder cable does not go to the drive’s ordinary terminals. It goes to an option card (often called a PG card or feedback card) that matches the encoder’s signal type. Check the card against the encoder’s output (TTL/RS422 line driver, HTL, or the absolute protocol) and supply voltage before ordering either one.
A simplified model for learning: it shows why feedback matters and how gain, encoder resolution and load change the result. Real drives add current limits, filters and slip compensation, so tune the real machine from its manual.
Counts per revolution = PPR × 4, because the drive counts both edges of channels A and B (quadrature). Resolution is not accuracy: backlash, belt stretch and gear errors add to the real error. If the encoder sits on the motor shaft, the load side is not measured.
Advantages
- Good position control without a servo motor
- Lower cost than a servo system for the same power
- High torque capability, up to large motor sizes
- Simple, rugged mechanics, standard IEC motors
- Suits long-travel axes
- Absolute and relative positioning modes in many drives
- Often built into modern VFDs
Limitations
- Lower accuracy than a servo: about ±0.1° to ±0.5° at the motor shaft is a realistic aim, against about ±0.01° to ±0.05° for servo systems
- Slower control loops (tens of milliseconds) and slower reversals
- An induction motor cannot stop instantly; deceleration ramps are needed, and the final approach is slow
- Cannot hold full torque at standstill as a servo can; use a brake for vertical or holding duty
- Backlash and slip in the mechanics limit the real accuracy
- Needs an encoder and a drive with feedback and tuning
Choose a servo instead when
You need accuracy tighter than about ±0.1°, cycle times under about 200 ms with fast reversals, full torque at zero speed, or a backlash-free transmission. A servo costs more but it is built for these jobs. See Motor Control Technologies.
Closed-loop VFD works well when
Accuracy of half a degree or looser is fine, moves are not too frequent, the motor is large, and the load is a rotary table, long conveyor or lift where a servo would be costly or hard to scale.
Use a brake for holding
An induction motor does not hold a vertical load for ever at zero speed. Use a mechanical brake, applied after the drive has stopped the motion, and release it before the drive starts to move.
| Control method | Speed regulation | Position use | Feedback |
|---|
Speed regulation is how closely the motor holds the set speed under load changes. Only the closed-loop methods can hold a position. For the full side-by-side of V/f, sensorless vector, closed-loop vector and servo, see VFD Control Methods.
| Parameter | What to set |
|---|
A typical move is a trapezoid: speed ramps up, runs at the set speed, and ramps down so the motor arrives slowly. Once the position error is inside the positioning window, the drive reports “in position”. A wider window settles sooner; a tighter window gives a more exact stop but may hunt or take longer. Ramp times of a few seconds suit large inertias, and shorter ones need more current.
Accuracy figures are typical and indicative; they depend on the encoder, drive, tuning and mechanics. Always follow your drive manual for parameters, feedback card, and wiring, and your site rules for electrical safety.