Final Design Overview

Navigation Menu

Description

How the System Works

System Components

Electrical Components

Electrical Schematics

Saftey Systems

Testing

Description

The final design comprises a converging device, where the shells are first poured into. From this point, the shells travel one by one to the "timing device". Using a DC motor-driven Geneva drive, the shells get pushed into a funnel to send into the rest of the system with a 4-second spacing between each other. At this point, the flow is not sufficient to push the shells through to the end, so another pump is attached to the system creating a sheath flow. The sheath fluid (water) creates a laminar flow that surrounds the core solution and brings it to the next portion of the system which is the sensor. We are using a Keyence fiber optic through beam photoelectric sensor which detects whether a shell is passing through the tube or not. This is meant to trigger the "camera" or in our case an LED light to inspect the shell to determine whether it should go in the "good" or "bad" batch. Once it passes through, the sorter, or what we call, the diverter, will push the shells in either the good or bad batch.

How the System Works

Step 1: Shell Loading & Flow Control

Shells are poured into the converger, flowing one by one into the timing device.
An aquarium pump maintains water levels, preventing clogging and ensuring shells remain submerged for continuous flow.

Step 2: Transition to Timing Devic

Aligned shells pass through a spigot onto the timing device.

Step 3: Timing & Dispensing

A rotating plate with evenly spaced holes controls shell release.
The spigot and plate geometry ensure only one shell per hole.
A DC motor with geneva drive rotates the plate, dispensing one shell every five seconds for inspection.

Step 4: Inspection Trigger & Imaging

A photoelectric sensor detects incoming shells, and an Arduino calculates delay for precise timing.
An LED flash simulates camera automation (future AI integration planned).

Step 5: Camera Window & Quality Check

A glass-paneled camera window allows sharp imaging.
Future AI software by General Atomics will determine shell quality.
Currently, manual green (pass) or red (fail) buttons control sorting.

Step 6: Sorting & Water Recycling

A servo-driven diverter rotates 180° to direct shells to good or bad sumps.
Shells are caught in mesh sumps, while excess water drains into a reservoir for recirculation, reducing waste.

System Components

1) Converging Component

2) Timing Device

3) Sensor

4) Imaging Window

5) Diverger

6) Pass/Fail Sumps

Converting Mechanism

Component: Angled pyramidal hopper for controlled shell dispensing
Function: Takes in a large beaker of water and foam shells, releasing them in a single-file line.
Moisture Maintenance: Remains semi-wet to keep shells submerged and prevent sticking.
Design Benefits: Uses a water pump to introduce a small amount of water to prevent clogging, and maintains a steady flow into the system

Timing Mechanism

Intermittent Rotation: The timing mechanism ensures controlled, periodic movement.
Rotation Ratio: The driven wheel rotates 1/20th per full driving pin rotation.
Shell Frequency: 1 shell every 4 seconds for consistent processing
Optimized Design: Ensures efficient shell separation for accurate sensor detection.
Flow Rate Calibration: Experimentally measured water output to maintain a constant system fill level.

Sensor

Sensor choice: Keyence Fiber Unit FU-18, a fiber optic through-beam photoelectric sensor.
Function: Emits an IR beam; light intensity changes when an object passes through.
Detection: In laminar flow, detects significant light intensity drops for all shell sizes.
Trigger: Camera activates when light intensity falls below a set threshold.
Velocity Calculation: Uses shell diameter and time in sensor range to determine trigger delay.
Signal Output: Amplifier sends an NPN signal to an Arduino, activating a relay for system components.

Imaging Window

The imaging window features orthogonal glass panels, allowing imaging from two different directions for a more comprehensive view.
Microscope slides are used for high optical clarity, ensuring precise and sharp imaging.
The photoelectric sensor triggers image capture at the optimal moment.
Future AI software by General Atomics will analyze shell quality automatically.

Diverging Mechanism

Component: Rotating tube insert with slant for flow diversion.
Function: Redirects shells without physically damaging the shells, ensuring safety.
Water Sealing: Diametric magnets maintain a watertight seal.
Gear Reduction: 2:1 ratio to match shell speed.
Motion Control: Connected to a servo motor, which rotates 90 degrees via PWM signals from the servo driver.

Electrical Components

Electrical Schematics

Main Components

Arduino UNO

Detection

Fiber Optic Photoelectric Sensor
5V LED

Safety

Float Switch
Emergency Stop

Actuators & Power Control

12V Relay
5V Servo Motor
5V DC Motor
12V Water Pump
12V-5V DC Converter
Momentary Push Buttons

Inside the electrical box

Control panel comprising of (from left to right) "good" and "bad" buttons for servo motor actuation, on/off switch, and emergency stop.

Saftey Systems

Failsafe Features: Implemented safety components to prevent system failures.
Float Switches: Placed in the converging mechanism and funnel after the timing mechanism.
- - Detect excess water flow and automatically cut power to the pumps.
  - Prevents overflow and potential damage to the system.

Emergency Stop Button: Installed in the control panel for compliance with safety guidelines.
- - Immediately shuts off power to the entire system when pressed.
  - Allows operators to quickly respond to emergencies or malfunctions.

Float switch turns off flow when contact is made

E-stop effectively turns off the system

Testing

Test Conditions

The test conditions of our experimental setup for the prototypes were as follows:

The flow conditions were constant after we impregnated the system with water and allowed the pumps to reach steady flow.
Once shells were inserted, we held the pump flow rate to 1 ft/s which matched the rate that the reservoirs released.
The shells we first used were polystyrene balls 4 mm in diameter, which are denser than the actual shells and sink in water. This allowed for the easiest test and preliminary proof of concept.
Later we tested with the actual shells, 3-4 mm in diameter. These shells closely matched what the operator would be testing with GA.
The tests were airtight inside of the system, and open to the atmosphere at the insert and exit of flow.

Prototype Results

When testing the first component, the converging mechanism and timing device, our hopper clogged on average 50% of the shells if the shells differed from the designed size by more than 33%. If the shells were oversized, they wouldn’t fit through the hopper exit chamber, and if they were too small, they were susceptible to piling up and clogging that way. For example, if the design was intended for shells with an outer diameter (OD) of 6mm and 4mm shells were fed, clogging was imminent. However, inserting shells with similar dimensions to those designed for lead to much less clogging and more effective convergence. This led us to narrowing the scope of the project to only testing shells within a 10% dimensional variance to the largest size it was designed for. We built two separate hoppers for sizes 3-4mm and 5-6mm, along with their respective rotating plate sizes. Since they are all modular and attached to one another, they are easy to swap out for the operator.

Final Design Results

As seen in the graph, the final system should perform as intended with foam shells up to 30 at a time. Testing of the Polystyrene shells shows that with further optimization of flow in the hopper, the # of shells to be tested at once could be increased.

Page updated

Report abuse