Eagleflyer Project
The Eagleflyer board is a very versatile device used as a controller and data collection device for amateur rocketry or remote controlled vehicles. Among its capabilities are a wide array of sensors and components, easy re-programmability, the ability to send/receive data real time to any PC hooked up with a HAM radio and the ability to be expanded with even more analog or digital sensors.
Sensors and Components
GPS Interface
1 MB of data storage
temperature sensor
barometric sensor
50G accelerometer
2 single-ended analog inputs
2 differential analog inputs
magnetic field sensor
4 high current mosfets
Easy Re-programmability
This board can be reprogrammed easily with a reprogramming adapter, on any Linux or Windows system with the free avr-gcc compiler (GCC cross-compiler). All code for the board is available on this site.
Radio Interface
When hooked up with a HAM radio the Eagleflyer board has the ability to send/receive data real time to any PC that is connected to a HAM radio within transmission range. This data transmission is done through the on board 1200 baud radio modem.
Expandability
The Eagleflyer board has room for even more sensors, making it extremely versatile. It has 2 separate 1X to 1000X adjustable gain differential analog inputs as well as standard SPI, I2C and UART digital interfaces as well as several multipurpose digital lines for custom interfaces.
What can the Eagle Flyer do exactly?
Some possible applications for this device would be to send telemetry data from a model rocket or remote-control plane, it could also be a great upgrade to your remote control boat or car, or it could even be used to control an autonomous robot
This board was made to work as a general purpose controller board for amateur rocketry. It has a magnetic sensor that detects the earth's magnetic field which is useful for magnetic apogee detection (known as "MAD", invented by Robert Galejs in 1999), it also has a 50G accelerometer, a barometric pressure sensor, an ambient temperature sensor, a 1200baud radio modem interface, 4 high current optoisolated N-channel mosfets, 1MB of flash memory storage, 2 separate 1X to 1000X adjustable gain differential analog inputs, a GPS interface, and an atmega32 AVR microcontroller to control the whole mess. The board measures approximately 5.5" by 1.7" (14cm by 4.3cm)
Software for the board
main software for the board (avr-gcc C code)
software for using the board as a teststand loadcell interface (avr-gcc C code)
software for using the board as a MAD (Magnetic Apogee Deployment) sensor (avr-gcc C code)
software snippets for using the board in different ways (R/C servo control, ignitor interface etc) (avr-gcc C code)
Board Pictures
pictures of the version 1.0 board
Board Documentation
document with more board photos, schematic, bill of materials (BOM), and board layout graphics (warning 9MB!)
datasheets (for chips that are used on the board)
Here is an excel file for the BOM (same info as in above pdf)
gerber files for the board - these are the files that I sent to the board manufacturer Olimex, 8 of the boards fit on their "DSQ" 320mm x 200mm double sided pcb panel, and costs $104US.
Here is the info for what the 8 gerber files in "gerber
files.zip" mean:
*.cmp Top, Via, Pad Component side
*.sol Bot, Via, Pad Solder side
*.plc tPl, Dim, tName, Silkscreen comp. side
*.pls bPl, Dim, bName, Silkscreen solder side
*.stc tStop Solder stop mask comp. side
*.sts bStop Solder stop mask sold. side
*.drd Drills, Holes Drill data for NC drill set
(also the *.drl file lists the drill sizes used on the board, and this is generated by running drillcfg.ulp)
Another way to write the layers: (this makes more sense when you look in the Eagle cam processor)
cmp (1,17,18) component side
drd (44,45) drills, holes
plc (20,121,125) silkscreen component side
pls (20,122,126) silkscreen solder side
sol (16,17,18) solder side
stc (29) solder stop mask component side
sts (30) solder stop mask solder side
note: to generate layers 121,122,125,126 silkwidth.ulp and silkwidth.scr (generated from silkwidth.ulp) must be run
Known Bugs With the First Version of the Board
1. The magnetometer chip is missing a 1uF capacitor used for resetting it. I added the capacitor to the schematic and soldered a surface mount capacitor on the boards and it works ok now.
2. One of the mosfet power traces is too thin on the board - the trace labeled "GNDA" on the schematic which connects mosfet T5 to T3. I soldered a copper wire on the board over this trace to increase the current carrying ability, but I haven't updated the layout for this bug yet.
3. The magnetometer slowly loses accuracy. I think this is because of the 5Volt "powerplane" trace near it on the board. This trace probably only carries a few ten's of milliamps but that may be enough to interfere with the very sensitive magnetometer chip. To fix this the layout would have to be changed by moving the magnetometer away from the power traces. An interim fix is to periodically reset the magnetometer every couple of seconds to keep it accurate. This seems to work fine in the software, and the magnetometer should be fine to use as a Magnetic Apogee Detector (MAD).
Mailing List
"eagleflyer-chat" is a mailing list for discussion about the EagleFlyer board and any related electronics/software/rocketry stuff! :)
Links
AVR-GCC cross compiler (avr-gcc/winavr) (runs on linux and windows)
Contact
contact me (remove "NOSPAM" from the email address)
check out my amateur rocketry website at www.rocketresearch.org