BLDC motors, is the free run (no load) current equal to the min current

[Infraviolet]

A simple question, but I'm not sure if my conclusion is right:
Is there a minimum current in the windings to make a BLDC motor move at all, I would assume there must be because having now bought a few exmaples for use in that low current BLDC driving project I'm planning, I've noticed that there is an amount of force needed to turn the rotor when the motor is unpowered, simply due to the attraction of the magnets in the rotor to metal parts in the stator. This comes in little bursts with the angle, so I guess it is also responsible for the "cogging" torque which remains present for a motor even when driven sinusoidally.

A motor's no-load current occurs when the only torque it is having to provide is that needed to overcome internal mechanical resistances. And at any speed the motor has some voltage across the coils, some back-emf opposing this voltage, and a current flowing which is related to the suppliedVoltage-backEMF, although with short pulses of voltage the inductance means it isn't obeying I=(suppliedVoltage-backEMF)/R but rather I=(1/L)*(suppliedVoltage-backEMF)*t where I is the current at the time (t) after a pulse of power begins being applied, with R being neglected.

The maximum speed with no load, for some voltage, is achieved when the back EMF gets large enough to limit the current to the no-load current. At this condition the current present is exactly enough to overcome the mechanical resistances in the motor, but there is no torque left to accelerate it to higher speeds. The same point would be reached, albeit at different speeds, for any voltage supplied (so long as not too small), and at the same no-load current.

Would this mean then that starting from rest the motor won't be able to begin moving until the same level of torque provided at the no-load maximum speed is present? In which case, for currents under the no load current there would never be enough torque provided to overcome this mechanical resistance?

Or does the torque needed at speed in the no load condition depend more on dynamic (proportional to shaft speed, or shaft speed squared...) friction effects rather than the same constant (would appear in equations without multiplication by a shaft speed term) mechanical resistance effects at low speed? In which case depending how serious the friction got at high speeds the no-load current at high speed could be uch larger than the minimum current neede to start at low speeds.

Thanks

[Siwastaja]

Yes, you are right BLDC motors have "cogging torque", or permanent magnets interacting with the iron core. Additionally, bearings have friction losses. As 100% efficiency is impossible, it takes non-zero torque to rotate a motor, and given no current, a running motor loses speed.

You also seem to have correct understanding that there must be voltage excess over the BEMF voltage to actually drive current to the motor, simply dI = dU/R. So given fixed DC bus input voltage, maximum current (and torque) that can be generated starts at excessively high value at zero RPM (so it needs to be capped to a sane value) and linearly drops, until the point where the available torque equals to your mechanical load and no more RPM can be gained; with no external mechanical load, that would be the internal mechanical losses of the motor.

This is ignoring the field weakening strategy which basically changes the RPM-to-voltage relationship (back-EMF constant) to allow higher speed.

[johansen]

Yes the minimum current will be required to overcome the cogging torque.

But once it starts moving its own inertia will carry it over., so the peak current will be lower if the drive has any mechanism to match the drive current to the back emf.

You could easily have a lot of excess current that is not needed.