# Building BLDC ESC - question on PWM duty-cycle vs motor speed

I'm currently working on a simple ESC design to drive a small BLDC motor (6A continuous max). I'm at a stage where I can start the motor reliably, but haven't yet achieved closed-loop control using zero-crossing. I have done some experiments where I keep increasing the commutation frequency, but at some point the motor stalls. The PWM duty cycle doesn't change as I increase commutation frequency.

My question is, does BLDC motors require higher current (increase in PWM duty cycle) at higher rotation rates? This is my intuition as higher current results in stronger magnetic field, which results in "faster" rate of change in angular velocity.

I want to spin the motor at higher frequencies reliably, that's all.

To clarify some of the terms,

Commutation Frequency - how fast the commutation steps change (1->6). This leads to changing speed in BLDC motors

PWM frequency - PWM signals are given during each commutation. Varying duty cycle here simulates varying supplied voltage to motor (and hence current)

## 1 Answer

A BLDC motor is similar to other motors in that the torque produced is proportional to the current, and the speed is proportional to the applied voltage (assuming commutation is done correctly.) Of course there are second order effects, but that's the basic idea.

So if your motor was ideal and lossless, running at high speed with no load would be zero power output and would require no current. With a real motor windage and other losses increase with speed even at no load, so there is more current required to overcome those losses.

The first-order increase in speed comes from an increase in effective applied voltage, which comes from appropriate commutation and PWM duty cycle.

• this is a rather clear explanation. So in order to achieve close-loop control, set target speed as a constant (commutation frequency), measure zero-crossing and vary PWM duty-cycle sounds like the logical thing to do from your explanation. I did notice (from playing around with parameters) that even at relatively low speed, I can control the current drawn by motor by varying PWM duty cycles. In this case some Power is wasted then (motor did heat up)? – user1906583 Jan 28 '15 at 22:43