Controllers - do they only manage duty cycle?

Hi everyone.
I've been reading up on BLDC controllers in preparation for a motorcycle build, and would appreciate it if anyone can confirm my understanding of these things.

As far as I can tell, all BLDC controllers use pulse width modulation (PWM) to control motor phase current.
And regardless of whether it's controlling the stator phase or the amplitude of the current, it's still just a function of duty cycle.

Now each time the controller switches a phase ON, the current through the phase winding is determined by I = V/Z, where
V = battery pack voltage, say 72V
Z = circuit impedance (resistance, capacitance and inductance)

If we only look at resistance, we can include phase winding resistance, battery internal resistance, MOSFET junctions, leads and connectors, say 0.2 ohm, there would be a short circuit current (SSC) of 360A (provided the pack could suppy this).

If I had set the motor controller to have max phase amps of 180A, it would effectively never exceed a 50% PWM duty cycle.
Now if my throttle was linearly mapped to motor torque, I would expect a 10% throttle to equal a 5% duty cycle (@ 360A SSC).
And a 100% throttle would equal a 50% duty cycle (@ 360A SSC).

Considering the simple case above, and ignoring inductance and capacitance and multiple phases for the time being, can anyone confirm that the controller set up and limits are really just functions of duty cycle at the end of the day?

Hi,
All that the controller can really do is switch. That is to turn on and turn off the switching devices, mosfets, IGBTs, or whatever. Everything else is just figuring out when to do it.

And I = V/Z is valid at stall, otherwise you must include a generated voltage in the equation.

Regards,

major

As major said the back EMF of the motor has to be factored into the equation and this results in you needing more and more system voltage to get higher motor rpm.

With regards to throttle response it depends entirely on how the system has been designed.

Some will work in a way where the speed of the motor is proportional to the load applied and voltage supply of the system.

Others will work in a more regulated fashion where the throttle actually defines a speed of rotation, rather than a % of maximum. In this case if the throttle response says spin the motor at 1krpm the controller will vary the MOSFET switching in an attempt to main that rpm within the limits of the specified current limit and available voltage headroom over the back EMF generated.

I guess there would be one more type of control which would be a torque control response to the throttle. Essentially the throttle would define a current limit. 10% throttle might = 10 amps, for example, and if the motor load is light, say you're going downhill, then it will spin quickly. Then if you start going up hill the load increases and the motor slows down. Here no attempt is made to keep the rotational speed constant it simply spins as fast as the current limit allows within the specified maximum set for max allowed motor rpm.

It all depends on how the controller works and how it's configured.

There are many ways and levels of detail to look at what is going on.

With a simple trapezoidal controller the first issue is which of the three pairs of motor conductors to apply voltage to, and then in which polarity. That's six possible voltage/polarity combinations.

When the FETs conduct and apply voltage the maximum current is determined by battery voltage minus back EMF divided by circuit resistance but the inductance limits the current rise according to L*di/dt (effectively adding to the motor's back EMF). So the current rises limited by the inductance, and the PWM duration determines how far it gets up the curve before the FETs stop conducting and the PWM cycle decay begins.

For a sinusoidal controller a more complex calculation is done and PWM controlled FET switched voltage is applied to two motor conductors producing currents in all three, the sum of which must be zero. A smoothly rotating magnetic field is produced instead of one that jumps 60 degrees with each commutation state. Two PWMs are generally used in this case.

So yes, controllers manage duty cycle and timing/direction of the voltage applied to the motor, but in a fairly complex manner.

Thanks guys! appreciate the informative responses!