 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
A great deal of effort went into the design of the electrical circuitry required to run the automated slot car controller. The diagram below illustrates the overall system.
move your mouse over the system elements
above and follow the links
The four components listed on the left were purchased off the shelf and integrated accordingly. The three on the right were designed and hand built in-house. For a complete description of the user interface that operates the system, see the Features section.
Motorola MC68HC11-E9 Microprocessor Unit (MCU)
The Motorola MC68HC11 microprocessor serves as the brain of the system. The E9 variant is effectively a 2MHz processor with 52 pins for I/O. It includes an internal clock, timer, A-D converter, 512b of internal RAM, 512b of internal EEPROM, and 5 I/O ports. See illustration at right for a block diagram of the MCU components.
This project utilized the following capabilities of the HC11:
- Input Capture: sensor pulses from cars
- Main Timer: overflow count and output compare for duty cycle control
- Parallel I/O: sensor position pulses, driving voltage measurements, toggle/dial/momentary switches (input), PWM control, LCD text display (output)
- Analog-to-Digital Converter: driving voltage and difficulty level
- Maskable Interrupts: timer overflows, reset race button
The Axiom EVB contains the Motorola MC68HC11-E9 64-pin microprocessor, and connects it with several peripheral devices for expanding memory and interface capabilities. In the Automated Slot Car system, pin assignments were as follows:
Inputs:
PE0: Auto Car Voltage
PE1: User Car Voltage
PE2: Difficulty Level Switch
PD4: Computer Control
PD5: Start Button
IRQ: Reset button
PA0: Starting Line sensor for auto car
PA1: Both into corner sensor for auto car
PA2: Out of corner sensor for auto car
PA3: Starting line sensor for user car
Outputs:
PA6: PWM to Auto Car
LCD Port
For more information on the EVB, see the Buffalo
Manual.
A power supply was used that converted line power to 5, 12, and 24 VDC. The 24V output was regulated to 15V, which powered the motors in the cars. 15V was found to be the maximum that the cars could use before flying off the track.
Click here for manufacturer specifications.
A liquid crystal display was used to display text during the startup sequence, speed and lap time during the race, and winner / loser results at the end of the race.
Click here for manufacturer specifications.
The slot cars have a small pin underneath which rides in the slot on the track and keeps the car in its lane. Slot-type photomicrosensors were mounted in the track, which produce a continuous 5V at the output until the infrared beam that spans the slot is broken. The sensors were mounted so that when the car passed a sensor, the pin would pass through the slot in the sensor and break the beam. During the period that the beam is blocked, the sensor output drops to zero.
The sensor circuit was designed to capture these zero pulses and send a signal to the MCU that it could detect (falling edge). These edges were used to trigger interrupt routines in the software that increased or decreased the duty cycle--and therefore speed of the car--accordingly.
The circuit operates such that sensor pulses pass through an RC filter and on to a comparator before closing the loop to the input capture (IC) pins on the MCU. See the illustration at the right.
Perhaps the greatest challenge in this project came in protecting the signal from noise on its path between the sensors and MCU.
At racing speed, the cars draw around 300mA at 15V, through sliding pickups on the rails and clicking across joints. This generates noise spikes very close to the sensors, which can cause false triggering to the MCU and subsequent loss of control.
To remedy this problem, shielded sensor wires were used, low-pass RC filters were built into the circuit, and comparators were used to try to clean up the signal. Several iterations of changing RC values in the filters were necessary before the best cutoff frequency was found, which balanced between noise reduction and sufficient notch depth to trigger an interrupt in the MCU.
The power delivered to the automated car is controlled via a two-stage PWM transistor circuit. The variable duty cycle square wave from the MCU is sent to the first of the circuit's two stages, which consist of an Optocoupler and a MOSFET. See the illustration on the right.
Voltage pulses across the LED in the Optocoupler cause current to flow through it, which in turn causes current pulses to flow through the transistor side from the power supply, multiplied but yet decoupled from the MCU by means of an infrared beam.
The emitter of the Optocoupler in turn feeds current through a voltage divider at the gate of a MOSFET transistor. The voltage at the gate controls the source/drain voltage at the output, which controls the voltage seen at the rails of the car.
This circuit provides complete isolation and protection between the MCU and the car, while allowing sufficient power to flow from the power supply to the car motor. The speed of the car is adjusted by adjusting the duty cycle of the square wave coming from the MCU.
Except for a small time delay
due to the dynamics of the car, which is accounted for in the software,
the MCU can control the car with such precision that is very difficult
for a human opponent to defeat.
In addition to the LCD screen which displays real-time velocity measurements of both cars to the user, two trees of LED's on the front panel also provide a qualitative measurement of the speed of each car. These trees, one for each car, are driven by an array of comparators arranged in a ladder-type configuration. See the illustration to the right.
A constant reference signal is applied to the inverting input of each of six comparators, increasing between each level of the voltage divider ladder. The car's voltage is divided appropriately and sent to the non-inverting input of each comparator. When the voltage difference at the inputs to each comparator is positive, the output to that comparator rails to its supply value and excites its LED on the tree.
The result is a string of LEDs that illuminate sequentially with the speed of each car, giving the user a sense of how fast each car is moving.
See a photo of this circuit here.
Home | Introduction | Concept | Timeline | Mechanical Design | Electrical Design | Software Design | Features | Team Members | Sponsors | Links
Ryan Krauss / Lisa
Ellis / Joe Frankel
For problems or questions regarding this web contact [Joe
Frankel].
Last updated: 11/25/02.