From Albatross

Introduction

The hardware consists of two main PCBs,

• /Main Board
  • At the front of the plane (in the canard)
  • Contains the CPU, memory
  • Contains the sensors
• /Engine Board
  • At the rear of the plane,
  • Contains the power electronics, and engine controller

Notes

• Need to do memory profiling on existing board to work out if we can get away with 32MB of RAM. Single-chip 64MB parts are slightly hard to find in low quantities.
• The main PCB will be mounted flat i.e. in the same plane as the wings, so if we can use the IDG300 gyro, and one of the 3-axis magnetometers, we don't need any daughterboards! (We can use one IDG300 for roll/pitch, and possible one ADXRS150 for yaw if necessary).
• The communications module (i.e. Aerocomm/Maxstream/whatever radio module OR cellphone module OR satellite module?) will not be on-board.
• The GPS receiver should either be on board, or on a separate board near the antenna (i.e. smart antenna idea), not on a daughterboard. I think eliminating daughterboards wherever possible is a worthy goal.
• Still have an supervisory microcontroller, either an Atmel AVR or Philips LPC2xxx ARM.
• Hardware over the air on/off (auto/manual) of some kind. Probably servo switch over. Apeliotes Project has one designed with discrete logic.
• Much simpler power supply arrangement (say, just reg. 3.3V in + unreg. 7.2V in for 5V LDO, main power supplies on engine PCB)
• Main battery, alternator and power supply circuits on Engine Management PCB (see below)
• No expansion bus, no ethernet, no USB host, no CAN bus
• USB client with usbnet gadget driver instead of ethernet
• Smaller, lighter, (hopefully) cheaper board
• Individual connectors on board (rather than a giant loom). Maybe? A loom does have a lot of advantages in a production system, but is less devel.-friendly...
• No servo controllers on main board (they are on the engine control board)
• More LEDs on the board :-)