Title: Automotive Control Solutions
1Automotive Control Solutions
Automotive Control Solutions Russell Potter
CTO, President Alex Gutica - CFO Brian Nelson
- CTO
- The AF Optimizer
- - An ENSC440 project -
2Contents
- The ACS Team
- The AF Optimizer
- The 440 project
- In-car Demo
- Now
- The future
- Conclusion
- Questions?
3Automotive Control Solutions
- A cutting-edge development team
- specialize in control of automotive performance
through electronic air/fuel optimization - Appeal to owners of any car, particularly older
vehicles with simple electronic control
4Who Are We?
- User Interface Firmware Lead
- Russell Potter
- DSP Firmware Lead
- Alex Gutica
- Hardware Lead
- Brian Nelson
5Internal Combustion Engine
- Requires a correct mixture of fuel and air in
order to function - Fuel is mixed with the air, compressed, and
ignited. - When ignited, the air/fuel mixture drives pistons
down, which turns a crankshaft.
6Fuel Delivery
- The most efficient burn mixture has an 14.71
air-fuel ratio. - The lean condition
- The air-fuel ratio is too high
- Results in detonation, power loss, increased
emissions - The rich condition
- The air-fuel ratio is too low
- Results in reduced economy, increased emissions,
power loss
7Fuel Injection System
- Proper fuel delivery is electronically controlled
through a fuel injection and ignition timing
system
8Fuel Delivery
- Based on a 2-variable present map in the ECU
- Load/Airflow meter
- RPM
9The Problem
- Fuel maps and fuel delivery are designed for
compromise - By modifying fuel delivery and consequently
air-fuel ratios, improvements can be made - Performance
- Economy
- Improved Emissions
10Why Modify an Already Tuned System?
- Are the original engineers incompetent?
- No but, the original characteristics of the motor
can be changed - Performance enhancements
- Changing the amount of air\fuel flowing into the
motor - General engine wear
- Perhaps a different compromise is desired
11Current Solutions for Modifying Air-Fuel Ratios
- Modifying the computer new fuel maps
- Model-specific
- Costly
- Lack of user-specific ability to tune
- Standalone systems
- Complete computer replacement is very intrusive
- Requires extensive, expensive tuning
- Mechanical solutions
- Rudimentary
- APEXi SAFC
- Our direct competition
12The AF Optimizer
- Its functionality and potential market
- Its competitive edge
- Features
- System Design and Implementation
- Hardware and firmware
13The AF Optimizer
- What does it do?
- Recalibrates air flow sensor data, while
monitoring car - Allows for flexible tuning of air fuel ratios
- Moves to different location on original fuel map
- Monitors automobile sensors with real-time visual
display to users
- Why would one buy it?
- To inexpensively and safely optimize delivery of
fuel to their engine
14Target Market
- Customer needs to tune their fuel system
- Desires better performance
- Wants a simple, noninvasive install
- 29 Billion Dollar aftermarket part industry
- Our target demographic is young people
- With older cars
- Who demand an inexpensive, feature packed fuel
control system
15Compatibility
- Compatible with wide range of manufacturers
- Required
- Fuel injection
- MAP or VAF sensor
- 0-5V Scale
- Reality Older, simple computer is better
16AF Optimizer Advantages
- First and foremost, price
- Versatile for use on many different vehicles
- Easy to install and remove
- Un-intrusive to the vehicle
- Real-time monitoring
- Works on older cars
17Feature Overview
- Airflow tuning features
- Shift Light Features
- Monitoring Features
18Tuning Features
- Many Tuning points provides more tunability
- based on RPM and Throttle
- RPM Tuning
- 2000-8750 RPM - 250 RPM increments
- 75 to 125 - 1 increments
- Linearly Interpolates between tuning points
3000 3250 3500 3750 gt102
110 109 107 98 100 97 98
19(No Transcript)
20Tuning Features
- Throttle Percentage Tuning
- 2 Calibration Curves High / Low Throttle
- User defined - based on throttle thresholds
- Throttle Thresholds
- Low e.g. lt30 throttle
- High e.g. gt90 throttle
- Linearly interpolates between the thresholds
21(No Transcript)
22Tuning Features
- Overall 2-Dimensional interpolation
- RPM and Throttle are variables
23Shift Light Features
- 2000-10000 RPM in 100 RPM increments
- 5 Sequentially lit LEDs
- LEDs light up every 100 RPM as you approach your
desired shift point - Increasing brightness
- Example where shift light set to 5000 RPM
24Monitoring Features
- Real-time monitoring of engines sensors
- RPM
- Throttle
- Battery Voltage
- O2 Sensor Voltage
- Airflow and calibration monitoring
- Calibration
- Pre and Post Airflow Voltages
RPM 3250rpm Throttle 80
Battery 14V O2 Sensor 220mV
Airflow 106 Pre 2050mV
Post 2184mV
25System Overview
26AF Optimizer Hardware
- Part Sourcing
- PIC 16F Microcontroller
- Maxim 10-bit DAC
- Noritake 4-line x 20 character VFD
- 5Volt Regulator
- Input Circuitry
27Hardware Challenges
- PIC Microcontroller
- Need to service the display, inputs and DAC fast
- Fast speed for calculations
- DAC Accuracy
- Power Management
- I/O conditioning
- Noisy car signals, voltage scaling
- Creating a stable, fast analog output with DAC
28AF Optimizer Firmware
- Performed two functions
- Sample inputs, calculate, output
- Handle interaction with user
- Buttons
- Display
- Written in C
- High level functionality
- Easy writing, debugging
- Memory Processor Usage
29Firmware Challenges
- Debugging and Simulation
- Simulator has limited functionality
- PIC was new to us
- Timing
- We had strict timing demands
- Needed all three hardware timers
- Required very careful time management
30Final Product Performance 1
- Successful integration into the vehicle was
dependent upon system response speed - Response to a 16Hz sine as airflow input
(unrealistic, but illustrates system performance)
31Final Product Performance 2
- A more realistic response to a square wave
- 1.5 ms system delay
- Small capacitor used to eliminate discontinuities
-
32Integration Challenges
- Very smooth integrating into the car
- Research of sensor signals
- In-car signal testing with oscilloscope
- Great lab setup for proper simulation
- Start up and Connection issues
- Starter draws current and dropped the battery
voltage - FIX Cap and diode
- Bad connections with breadboard and car wiring
- FIX Soldered car wiring harness PCB
- Car has a bad O2 Sensor
33Demo Time
- Things to demonstrate
- Monitoring
- How to setup throttle values
- Shift light
- How to set airflow calibrations
- Car running and driving
- Out to B-LOT everyone
34Dyno Results
- Very Impressive Results!!
- Running too rich loses power.
- Running too lean loses power
- We are able to change the air fuel ratios
- Here are the results
35(No Transcript)
36Baseline Run
37Rich Run
- Tried 120 but threw Check Engine Light
- Tuned to 110 from 2000-6700 RPM (redline)
- Results
38(No Transcript)
39Rich Run
40Lean Run
- Running too lean lost power in low RPMs
- Original ECU runs too rich at high RPMs
- Use AF Optimizer to lean high RPMs
- Results
41(No Transcript)
42(No Transcript)
43Engineering Considerations
- Positive system feedback due to vehicle velocity
- Does not affect airflow into engine
- Dynamometer is appropriate for vehicle tuning
- System memory considerations (derivatives)
- Precautions were taken to prevent derivative
reversal - Airflow signal adjustability range limited to 75
- 125 - High and low throttle curves at least 20 of
entire throttle range apart - Testing indicates airflow signal changes much
faster than throttle
44(No Transcript)
45Budget
- Proposed Budget 665
- Actual Spending 200
- Difference -465
- Summary
- We UNDERSPENT!
- Due to building only 1 prototype, cheap dyno time
46AF Optimizer Schedule
Predicted Completion Dates February 27 March
6 March 13 March 24
Actual Completion Dates March 15 March 6 March
17 April 4
Firmware Hardware System Integration Final
Testing
47AF Optimizer Now
- Have a fully functional prototype!
- Weve priced many parts in quantities
- The display dominates
- Accounts for as much cost as all others combined
- Produced prototype PCB and casing
- Plans underway for production model
- Designed to use 1 PIC for cost
48AF Optimizer The Future
- Small distribution at first
- Use online car clubs for marketing
- Will hopefully get feedback
- If all goes well, incorporate!
- Outsource manufacture to overseas
- Build relationships with distributors
- Maybe make a few
49Future Design Considerations
- Split it up into modules
- Display and buttons
- Main PC Board and wiring to cars wiring
- Shift light (remotely mounted)
- Optimize PCB designs
50ACS Team Summary
- This was a great learning experience
- Conclusions
- Acknowledgements
51Lessons Learned
- Time and hard work invested early pays off
- Research allows for easier problem solving
- Documentation is important for complicated
projects
52Conclusions
- We chose a great project
- Because we love cars
- It was complex, but workable
- We honed our skills learned in 4 years
- This product can make money
53Acknowledgements
- Scott, Lucky, Fred
- Dave Atchison, for experience with PIC and Dyno
time - ESSEF funding
54Questions?