QuadRover and PropNIC SPI interface

NASA Robotics Team

I'm having a bit of a problem. Using the PropNIC on the Propeller dev board is fine. I'm able to get the UDP packet, decode it, compute heading, determine which way to travel, etc. We're at the point of using the radios for navigation, so I took the PropNIC and connected it to the Propeller board on the QuadRover (AUX pins 19-15). The PropNIC is able to see ethernet traffic, but my Propeller is not able to see that data. After a bunch of debugging I found that the 10k resistor in series from the AUX header to Propeller is causing the problems. This resistor is there so you can connect 5V devices to the Propeller. The PropNIC uses an SPI interface to send over the ethernet packet data. When I scope the SPI lines from the PropNIC to the Propeller, the wave form is stretched out (compared to normal operation). It looks like a RC charging / discharging curve. I took the PropNIC to the dev board and again tested it with and without resistors in series, and it never worked when the resistors are present. We also added a hex buffer on the lines to maybe try and isolate the lines, but that didn't help either. I don't know why the resistor would pose a problem, but it's doing something to the signal lines. Anyone seems this before or have any ideas of what would be causing the problem?

Thanks guys!

DJSandberg

NRT,

The only thing I can think of is a wire wound resistor acting like an inductor or choke coil? It would be a long shot. Make certain the 10K resistor is a carbon film.

Daryl

NASA Robotics Team

The resistor is a 10k surface mount component (0805 I think). I'm fairly certain it's a carbon film.

I found a work around by using the data lines going to the accelerometer, compass, and GPS. This is not the way we would like to use the device since we'd like to use the accelerometer.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
http://www.youtube.com/user/NASAROBOTICS