Playing with Motors
(Backdated: Oct. 25, 2016)
The goal of this experimentation was to try driving a motor that ran on 12 volts DC with an Arduino (at 5V) by low-side driving it with a large power FET. I scavenged around for motors for a few days in various stockrooms, and I settled on a large-ish brushed DC motor with a heavy cast gearbox on it. All told, it’s about 5 inches square with a solid half-inch stub shaft. I picked a 30V 17A MOSFET, which is arguably overkill, but I wanted to stay well within a margin of safety for experimentation. I programmed the arduino to run a PWM sweep on a digital GPIO pin that I attached to the gate of the FET. I included a pull-down resistor across the FET from gate to ground in order to eliminate a floating input. I also connected the motor to a benchtop power supply with some 22AWG solid-core hook-up wire through the FET on a breadboard. Obviously not a solution for high-current, but I figured since there wasn’t any load on the motor, it wouldn’t be a problem.
Revision 0 was able to run the motor just fine, although looking at the low side of the motor with an oscilloscope I noticed that the back EMF from the motor coil was spiking to +- 30V, which was troubling to say the least. For Revision 1, I stripped some insulation on the motor leads and added a flyback diode in parallel with the motor to allow the back EMF to spin down without damaging the FET. Unfortunately, in connecting Revision 1, I accidentally touched the low side motor lead to the GPIO pin on my Arduino and killed the processor chip. To get Revision 1 working (and to avoid further mishaps), I replaced my poor Arduino with a Ruggeduino. Revision 1 worked quite nicely as well.
Now that I have the motor working with no load, I’m starting work on Revision 2. Once the motor is under load, it will draw much more current than the 0.3A it’s drawing now. This will either start to heat up my hook up wires or treat my breadboard like an overly-complicated 1A fuse. The plan is to use 18AWG speaker wire and proto-board to make a more resilient version of Revision 1. I’ll have to make sure that I’m not routing too much current through solder bridges and perf-board vias, but ideally it should just be structure and the wire and FET will be dealing with all of the high-current pathways.
This will turn the drill bit in our Rev. 0 prototype, and hopefully it’ll be good enough for our final extruder!
