Thread Tools
This thread is privately moderated by Jack Crossfire, who may elect to delete unwanted replies.
Jun 26, 2014, 10:16 PM
Registered User
Jack Crossfire's Avatar
Thread OP

Brushless direct drive rover busted

A motor is tested for a winding error by using the right hand rule, a fixed current source, & permanent magnet. 10 & 3 were swapped.

The balancer was officially recycled after an attempt to make it balance. There was a lot more involved than expected. Mass distribution made a difference. Less mass was better. It couldn't balance with a battery anywhere on it. An outer loop needed to calculate a target angle based on the number of motor steps. An inner loop needed to hit the target angle by giving motor steps. It didn't seem to have enough traction on the carpet. There seemed to be a maximum limit on the feedback rate.

The balancer became a simple rover.

Direct drive worked, at very low speeds. Spinning the motors up in micro stepper mode, then sustaining them by driving 2 phases in commutation mode was the most effective way. That got up to a brisk walk. Sustaining them in stepper mode didn't produce as much torque as commutation mode. They didn't have enough torque to use back EMF mode at all.

Back EMF mode didn't really obey the reference design, because the working RPM was too low. A workaround was to lowpass filter the commutation time. This meant it took 10 commutations to respond to a change in load or voltage, while the reference design could respond in 1 commutation. As soon as it hit a bump, it stalled. Don't think the reference RPM would have had enough torque, either.

So the direct drive idea was busted. It would be terribly inefficient, even if it had enough torque. A transmission would be better. Would consider a propeller more efficient, but very large. It was a useful study in motor control, making brushless motors go both ways, transitioning between a very wide speed range, transitioning from micro stepping to back EMF mode, controlling 2 motors from 1 CPU, making a tiny rover go very fast.

All this motor controlling brought to mind how many steps it takes a quad copter to change the speed of a propeller. It has to send a PWM or I2C command, which causes another PWM signal to change the amount of current, which changes the crossing time of the back EMF. If the current changes after the crossing time is already detected, it has to wait 1 extra commutation. Eventually, the controller derives a new commutation time from the new crossing time. Later, the commutation is signaled, which actually changes the speed of the propeller.
Sign up now
to remove ads between posts

Quick Reply
Thread Tools