Taking control of the motors

The first step in this project was to get rid of the exsiting control logic, and make my own replacements. Because this was a relatively cheap car, this control logic was limited to full steam ahead or reverse and full steering left and right. This was done with two simple DC motors.

With the parts I had on hand, the best way to "upgrade" this was with an Arduino Uno mounted with a custom 3D printed bracket. I took power from the battery pack already built into the chassis, and used a L293D dual H-bridge to drive the two DC motors. Although I had no direct control when the car was set free, I could let it go to drive in circles.

While testing the car driving in circles, the car had a habit of starting to drift because I lost the rear tyres. While this was fun to watch, the plastic rims began to get seriously scratched up. While talking about this with my partner, they suggested using hot glue on the rims to make my own tyres. to do this, I tried to spin the wheels as slow as I could while applying a consistent amount of hot glue. As it turns out, the wheels didnt spin slowly enough and a lot of hot glue just got splattered onto me and my clothes. Oops. Fortunately, the tyres did work to give me all the grip I could ever need.

Using the 4s AA battery setup that the car had lead to some weak acceleration, so I removed the batteries from the holder and soldered on a 9V connector. To limit the amount of money I was spending on 9V batteries, I bought rechargable batteries. With this upgrade in voltage, the acceleration was almost scary. I haven't managed to find a 0-60 number, unfortunately.

(I will probably be misusing technical terms for steering throughout this section. Sorry)

Upgrading steering

The stock steering system was a DC motor that had a pinion gear that interfaced with a rack on the steering bar to ultimately make the wheels rotate. This obviously had upsides, like it being simple and cheap. It also had numberous downsides.

• There is almost no way to accurately control it from an Arduino. Applying \(\p, Vcc\) gives full lock either left or right, but using PWM didn't give fine control
• The motor would stall when it reached full lock. This meant there was no back EMF like there is in normal operation, drawwing a lot of extra current.
• It is HEAVY. Judging with my hands, the DC motor is about twice as heavy as a 9g servo motor. This obviously limits steering and acceleration performance.
• When the motor is stalled and drawing more current, the internal resistence of the battery will lead to the voltage sagging. This further limits the acceleration and top speed of the car as there is less poer going to the drive motor
• The extra power dissipation means more heat in the car. This isn't a major issue at the moment, but as I add more instrumentation to the car, this could become an issue.

To upgrade the steering system, I replaced this DC motor with a small servo motor. This gives me fine contol, as with all servos, and it will draw less current when at a desired steering angle. Additionally, I have many spare servos on hand in the event that it is damaged in a crash.

One of the worries I had when changing to this motor was that the forces on the motor at high speed may be hard for the servo to overcome. I think this is unlikely to be an issue because the chassis isn't complex enough to have positive caster, meaning there is unlikely to be significant forces on it. There is a bit of play in the steering system due to the resoution of a 3D printer and the need to have the steering slightly loose. This is probably small enough to be fine for the servo to handle