Webtrain Decoupling

Download Report

Transcript Webtrain Decoupling

Webtrain Decoupling
Adam Kadolph
EE451
Bradley University
Advisors: Dr. Irwin, Dr. Schertz
Design Review Presentation - Week 12
Tuesday 3/6
Outline
•
•
•
Introduction
Timeline
Weeks 8-12
– Sensor
– Test Programs
– Automatic Decoupling
•
The Current Problem
2
Introduction
• The goal of this project is to design a
system to automatically decouple cars
from the train.
• Recalling from last semester
– I began to research sensors
– Made a test program to decouple train cars
with a few train controls
3
Timeline
4
Week 8
• Test sensors and determine one most
suitable for webtrain
– Photocell
• Pros: Small, simple to use, very sensitive to light
• Cons: Slow switching time
– Phototransistor
• Pros: Efficient
• Cons: Narrow angle of detection, difficult to use,
would have to be mounted on the sides of track
5
Sensor
• CdS Photocell
– Resistance between 1k (light) and 100k (dark)
Ohms
– Fits nicely in the track (only 1 tie removed)
– Simple to use
• Determined voltages that correspond to
different situations with a train car
– Three states: nothing, coupler, and train
6
Week 9
Built Test Track with Sensor
Designed circuit for Photocell
7
Week 9
• Determined voltage thresholds for three
states (nothing, coupler, train)
• Coded program to measure and display
voltages from circuit
8
Voltage Thresholds
Nothing State: 0-0.28 V and 0.52-0.84V
9
Voltage Thresholds
Coupler State: 0.28-0.52 V
10
Voltage Thresholds
Train State: 0.84 to 5.00 V
11
Test Programs
• Sensorx (working)
– Program to test one photocell
– Display on LCD
– Used with test track
• Sensor4x (working)
– Program to test four photocells
– Display voltages on LCD
12
Week 10-12
• Coded a basic program to send DCC
signals to the train depending on the
measured voltages from the sensor
• Coded another program to measure
values from four photocells
13
Test Programs
• Autodecouple (not working)
– Exclusively for Automatic Decoupling
– Voltage from photocell circuit will change train
speed
• Decouple (semi-working)
– Manual Decoupling Testing
– Automatic Decoupling Testing
14
The Current Problem
• Timing issues with DCC signals being sent
to the train
– Test program is sending signal too fast
causing the Microcontroller to reset itself
– The problem may lie in the previous year’s
code used to create the DCC signals
15
Fixing the Problem
• The timing contraints of DCC signals
– ‘1’ bit has 58 us low and high time
– ‘0’ bit has >95 us low and high
• Use of Timer 2 to fix problem
– Pro: Advanced timing features allow faster
more efficient timing
• Previous Code uses Timer 0
– Con: Time spent setting up Timer
16
Questions?
The End
17