Drive Module

Preface

Because at the time the current iteration had issues with fail safe sequences, and motors had to be safely mounted in different orientations, I was assigned by the SJSU Robotics Club to create an iteration that resolved those issues.

Objectives

Own a drive module that was more structurally rigid, provides less load to the motors, able to rotate continuously, made installation and removal of motors easier.
Manage the weight as the module will only add weight from added complexity and components.
Create a layout that allowed easy installation of a limit switch that does not risk damage if the drive module overshoots.
Design and fabricate various 3D printed and milled parts that minimized weight while not yielding in extreme load cases.

Design Draft

Early concept for drive module, notice how the propulsion motor spins with the module, unlike the final iteration.

Later concept design draft (uses vertical bevel gears rather than timing belts)

This mechanism took many iterations of concept drawings to minimize uncertainty on its specifications. Since swerve drives of this size contains almost no references online, a lot of experimentation had to be done to successful design a viable module.

A ton of consideration had to be considered for the shafts, as a big concern was how the shafts would react as a cantilever to the wheel in the case the rover drove over rough terrain. In extreme cases, the rover might hit a big immovable rock, making the strengthening of the module an absolute priority.

Designing

Drive Module (Wheel designed by Maya Enriquez)

Currently, the drive module is undergoing constant revisions to improve structurally and in terms of weight. To also negate risk of cantilevering, a counter brace was added to the opposite end of the wheel for stress to be distributed. A ton of pocketing is also implemented to minimize weight, as this iteration of module has become much heavier

The biggest design block was figuring out a coupler design that fit the mounting pattern of the pancake motor while minimizing the cantilever effect on the shaft that connected the motor to the drive module.

As such, with the help and guidance of alumni and mentors, I implemented a thrust bearing that surrounded the coupler to remove wiggling and prevent stress concentrating only at the coupler part.

Initial draft concept for thrust bearing configuration (The coupler is called the hub mount here)

Beyond the iteration that was manufactured, the geometry has since been optimized, hollowing out the main structure by pocketing as well as removing unnecessary chunks from the side plates that the former version had.

To validate this optimization, I used Onshape Simulation to measure the FOS in extreme load cases, noting that the structure is attached to a fixed part rather than hanging from the rover’s suspension leg.

I was also assigned to change the steering motor to adapt to the new drive module configuration. As such, I proposed the Gobilda 30 RPM yellow jacket motor and validated it using hand calculations. I also gave the steering layout a lower to high gear ratio of 1:1.8 (60:108) to maximize torque, as the rover will experience conditions where strong steering is required. This is especially essential because module is heavier, meaning that steering experiences more moment and friction.

Calculation for steering configuration, includes load cases

Manufacturing/Assembling It

Because the rover needed wheel modules as soon as possible for presenting at events, placeholder 3D printed parts were created for the side plates and the motor mount.

Base structure without side plates, partially assembled

Unoptimized structure with 3D printed side plates

Steer gearing and limit switch mount implementation (The blue part that holds the thrust bearing contains a nub that triggers the switch)