Publication Date

Fall 12-6-2024

Document Type

Final Report

Embargo Date

12-5-2024

Problem Statement

The objective of this project was to design and build an automatic ball bearing sorter to sort bearings by both material (steel, brass, plastic) and size (½ inch and ¼ inch) into separate packaging stations. The system was designed to process up to 80 balls poured simultaneously into a hopper and accurately sort them in under one minute, with a target accuracy of 90% or greater. The customer, Dr. Potter, provided specifications and requirements for the device, including a compact size (≤ 20x20x20 inches) and weight (≤ 15 lbs). Additional customer needs included performance reliability, manufacturability, safety, speed, and ease of assembly. The sorter was built on an aluminum rail base frame, starting with a hopper that uses a slanted ramp with guiding rails to prevent jamming. After the bearings are sorted by size through the diverging rails, they are directed into two separate tracks—one for ½-inch bearings and one for ¼-inch bearings. Each track features a catcher area to catch the bearings as they depart the diverging rails, followed by a downward sloped tube where the bearings roll until they hit a vertical wall. A light sensor analyzes the material of each bearing and sends a signal to a solenoid, which launches the bearing over the wall and into one of three possible paths. The paths are managed by a servo motor that directs the moving tube piece, guiding the bearing to the appropriate packaging station according to its type of material. The device’s electrical components are controlled by an Arduino system, with all parts connected through a custom circuit and code to ensure proper signaling. Most mechanical components—aside from the frame and electrical parts—were CADdesigned and 3D printed. The device meets key performance goals, consistently adhering to size and processing time requirements, with an average cycle time of 40 seconds. Although the target precision was 90%, the current model achieves 85%, which falls short of the target, but still shows strong and impressive results, with room for fine-tuning and improvement.

Author's School

McKelvey School of Engineering

Author's Department

Mechanical Engineering and Materials Science

Class Name

Mechanical Engineering Design Project (MEMS 411)

Share

COinS