Sensor vs sensorless ESC starting torque

[Circlotron]

Just ordered one of those brushless 3-phase motor drivers that are commonly used in model aircraft, race cars etc. You can get ones for motors that have position sensors as well as sensorless ones that use the motor back emf for determining rotor position. The motor I have is about the size of a coke can and has three Hall effect sensors but I chose the sensorless type controller so I can try it out on any other motor too.

The question is, if sensorless controllers are available, why do some motors still have position sensors? Do they have greater starting torque because of not having to be initially jiggled before rotation begins? Or might that initial jiggling be unacceptable in some applications and a clean start is important?

[max_torque]

It mainly depends upon the load, rather than the motor.

For a propeller, the torque at zero speed is (effectively) zero, and you don't care if it initially jumps backwards a bit as the speed controller first fires fully open loop.  Because the torque is low at low speeds, an purely open loop commutation sequence is very likely to drag the prop along too, meaning enough speed can be gained to enable a high enough Bemf to be reached to enable the system to swap into sensorless closed loop control.

For a motor say driving a car along, that's rather different!  (there can be max torque at zero speed, and you don't want the car to jump backwards or forwards when you just want to inch forwards slowly......)

It also can depend upon the motor control strategy being used. Simple block commutated controllers, or voltage controlled sinusoidal voltage control ones can get away with basic phase position sensing, whilst systems running advanced strategies, such as FOC are likely to have physical position measurement systems for improved accuracy and failure effects management etc