# Induction Motor Rotor Position and Speed Control

I am not an expert in this, but I have built some basic BLDC motors. I was recently reading up on induction motors. While I understand the idea of the AC causing a rotating magnetic field, the part I don't understand is how the speed is controlled precisely in induction motors.

I know BLDC motors are controlled by an Electric Speed Controller (ESC) that changes the magnetic field orientations based on the position of the rotor. The position of the rotor is determined by a hall effect sensor or back EMF through the coils. From my reading, the speed of an induction motor is controlled by a Variable Frequency Drive (VFD). However, when I dig into the details, the result of this is a change in the speed that the field rotates which seems no different from running a BLDC motor without knowing the position of the rotor (i.e. open-loop). The problem being, that without knowing the position of the motor it is easy for the rotation of the field to get our of sync with the position of the rotor. This is easily seen with magnetic coupling as well, where permanent magnets on the rotor are being dragged around my paired rotating magnets. It works up to a point, but once the rotation of the magnets exceeds a certain speed, the system goes out of sync and the rotor stops spinning.

So, my question is - how is the speed of induction motors controlled precisely without knowing the position of the rotor? Or, put another way - why doesn't speed control in induction motors require knowledge of rotor position, but speed control in BLDC motors does?

Vector Control algorithms in the VFD track the rotor speed and make adjustments to the PWM pattern to maintain sped and/or torque by creating (through testing) a mathematical model of the motor in it's memory and comparing what the motor SHOULD be doing to what is actually happening. In a full Flux Vector Control (FVC) version, the rotor position is fed back to the VFD via a shaft encoder. But there is also what's called "Sensorless Vector Control" that does not need the shaft encoder. What is done to monitor relative rotor position is to have highly sensitive current sensors in the VFD that can detect the anomalies in the stator current that are the result of the rotor bars passing through the stator fields. So in SVC, you cannot get absolute rotor position, but you can track the effects of load changes relative to what the VFD is telling the motor to do, so that the vector algorithm can make the adjustments.

Or... you can leave the VFD is open loop V/Hz (Scalar) control mode and not know what's happening in the motor. For a lot of tasks performed by induction motors, this is fine. Centrifugal pumps and fans for example. Centrifugal pumps and fans make up around 70% of all AC induction motor applications and vector control is largely irrelevant, because the entire task is sloppier than what the motor does anyway.

• Excellent, makes a lot of sense - kind of how I thought it would work, but couldn't find any information about it. Thanks. Nov 12, 2020 at 4:02

Induction motor does not have any magnets on the rotor. It has windings. You can drive induction motor by AC sourse without any controller.

BLDC motor has magnets on the rotor and it must be driven by electronic speed controllers. Generally, BLDC works with PWM signals.

As you said, the rotor position can easily measured by some sensors on BLDC. However, on induction motor case it is not necessary actually. If you want to use induction motor with varying speeds not constant,you can change the supplied frequency of AC power to reduce or increase the RPM of the induction motor.

I think it is just about usage areas and physical properties of these two motors.

• Thanks, yes I was aware that the rotor of induction motor has windings not magnets, but fundamentally they both seem to boil down to a magnetic (or magnetised) rotor following a rotating magnetic field. Increasing the frequency of AC motor seems the same as increasing the speed of the open-loop phase changes in BLDC. In AC motor, surely you can't start accelerate the frequency too quickly without stalling. I am missing something about why positional knowledge (and feedback) is required in one, but not the other. Are VFDs just very will calibrated to the motor? Nov 2, 2020 at 9:21