Motors Update: Encoders, MOSFETS and Airplane Propellers

You would think that getting a motor to move a car forward would be a fairly straightforward process. Unfortunately, like many things in life, it’s not that simple. This year’s Shell Eco-marathon wrapped up in April 2015. Since then, our motors team has been hard at work trying to get the motor that moves our new car, Alice, to spin smoothly. As it so happens, this particular motor was intended to spin hobby airplane propellers and so there are a few obstacles in the way to say the least.

This slideshow requires JavaScript.

The biggest issue right now is that the default setting on the motor utilizes components known as Hall-Effect (HE) sensors. These sensors communicate important information about the motor’s current state to the motor controller. We weren’t sure if the HE sensors would work well given the small size of the motor and so decided to get the next best thing: an encoder. The motor encoder accomplishes the same goal as the Hall-Effect sensors i.e. letting the motor controller know about the motor’s current state. When the three components work perfectly, the encoder provides control over the speed of the motor to the car driver. It’s the difference between gently accelerating and shooting off in a potentially explosive roller-coaster.

At present, the motor does a bit of a jig and then shorts the circuit. This could possibly be solved by finding the right set of parameters for the controller. Getting those parameters is, unfortunately, not a trivial task: We are in communication with RoboTeq, the company that manufactured the motor controller, and are working with their engineers to sort things out.

There is a silver lining to this however, the old light bulb errr… LED clicked and we realized that it would be possible to back calculate the necessary parameters from the max RPM and number of paired-pole magnets in the motor.

This slideshow requires JavaScript.

The only thing stopping us from testing this theory are a couple of blown MOSFET drivers on the controller. What happened was, while trying to get the motor to work, the motor was spun by hand which created more back EMF than the MOSFET drivers could handle. Things were not looking good as our controller was mostly useless at this point.

Fortunately, we had a backup plan: using a component known as an Electronic Speed Control (ESC) in place of our motor controller. The beauty of this component is that it doesn’t need feedback from the encoder; it uses back EMF to get information about the motor’s current state. We just finished setting it up and plan to start testing soon.


Electronic Speed Controller (ESC)

The bottomline is, we wanted to use RobotTeq’s motor controller but compatibility issues have forced us to get an off-the-shelf ESC instead. The good news is that we have a potential way to get the motor spinning with minimal effort. Plus, we now also have a better idea of how we can go about designing our own motor controller when the time comes.

For the future, the goal is to get the motor running and see whether it has enough torque to push the car forward from a cold start. As mentioned before, this motor was intended to spin propellers in hobby airplanes which is probably why the max torque isn’t documented anywhere. To obtain this crucial data, we plan to connect a dyno to the motor and run it at different settings. This should give us more information about how much torque the motor can generate and under what conditions the motor performs most efficiently.

That’s the current state of affairs as far as the motors department is concerned. If you are interested in joining the team, please do not hesitate to speak to any of our members in person or send us an email at! Until next time folks.


About Adnan Arif

A fourth year Chemical Engineering student at the U of A, Adnan is passionate about alternative energy, climate change and computer programming. Between assignments, exams and killing people in GTA, he tries to find time to work on EcoCar's Electrical and PR teams.

Comment is closed.