The No-Op Tank | Results

N64 Controller

Interfacing with the N64 controller was by far the most difficult part of this project. Initial electrical problems led way to timing issues and invalid responses. The initial circuit design that was used included a resistor from the transistor to ground. This resistor is not necessary, and causes the controller to not respond.

Timing issues when reading the controller response were the next major hurdle. Attempting to sample based on the expected timing proved to be too unreliable. To sample reliably, the hardware must sync with the falling edge of each bit. Occasionally, the controller fails to respond, or responds with a 24 bit sequence instead of a 32 bit sequence. Our solution to this problem was to do error correction in hardware to avoid these problematic signals.

Ultrasonic Sensor

The functionality of the ultrasonic sensor was greatly compromised by the noise and vibration produced by the tank drive motors, and the load spool motor on the disc shooter. The noise caused the sensor to respond sporadically, and caused the distance measurements to be wildly inaccurate. To solve this problem, distance sampling is left up to the user. It should only be initiated when all of the motors are off.

Servo and Infrared Sensor

The servo and the infrared sensor work when used independently, but cannot function properly when the higher powered motors are being used. This is probably due to the inductance of the extremely long ribbon cable used to connect all of the tank components to the board. The close proximity of the wires on the ribbon cable makes this issue worse.

Tank

The IO expansion board did not supply enough power to drive the tank. The speed of the tank was very slow and not acceptable. It came to a complete stop if the two treads were run in opposite directions. The MPC823 power supply was used in conjuction with a 9 V voltage regulator to provide adequate power.

LCD

We were also unable to use the LCD display at the same time with the rest of the components due to the lack of control pins. Pins 9-16 are mapped to the ADC, therefore not allowing us to use them for the control signals for the LCD display as we originally intended.

EECS 373 Final Project Winter 06 - Results




	Navigation	N64 Controller Interfacing with the N64 controller was by far the most difficult part of this project. Initial electrical problems led way to timing issues and invalid responses. The initial circuit design that was used included a resistor from the transistor to ground. This resistor is not necessary, and causes the controller to not respond. Timing issues when reading the controller response were the next major hurdle. Attempting to sample based on the expected timing proved to be too unreliable. To sample reliably, the hardware must sync with the falling edge of each bit. Occasionally, the controller fails to respond, or responds with a 24 bit sequence instead of a 32 bit sequence. Our solution to this problem was to do error correction in hardware to avoid these problematic signals. Ultrasonic Sensor The functionality of the ultrasonic sensor was greatly compromised by the noise and vibration produced by the tank drive motors, and the load spool motor on the disc shooter. The noise caused the sensor to respond sporadically, and caused the distance measurements to be wildly inaccurate. To solve this problem, distance sampling is left up to the user. It should only be initiated when all of the motors are off. Servo and Infrared Sensor The servo and the infrared sensor work when used independently, but cannot function properly when the higher powered motors are being used. This is probably due to the inductance of the extremely long ribbon cable used to connect all of the tank components to the board. The close proximity of the wires on the ribbon cable makes this issue worse. Tank The IO expansion board did not supply enough power to drive the tank. The speed of the tank was very slow and not acceptable. It came to a complete stop if the two treads were run in opposite directions. The MPC823 power supply was used in conjuction with a 9 V voltage regulator to provide adequate power. LCD We were also unable to use the LCD display at the same time with the rest of the components due to the lack of control pins. Pins 9-16 are mapped to the ADC, therefore not allowing us to use them for the control signals for the LCD display as we originally intended.
	Introduction High Level Design Member Tasks Hardware Design Software Design Results Conclusions Media