RPi Interface Rev B Test Plan

Intro

Only able to assemble 2x because of limitted budget. One for peggy, one for houston is POR
testing plan developed and executed @Daniel Puratich just meant as a minimum statement of what will be tested
- assistance from software people required to setup rpi and pixhawk
proof of completion of these as described in this doc is given in RPi Interface Rev B Photos
Testing the “Competition 2025” Variant only because this is the only one we’re using for now.

does buck work on its own
- yes
- no rails are shorted (probed in advance)
- reasonable power supply current
does hot plugging with battery kill it
- doesnt die
- how does it look onscope?
  - looks good, see photos
Does it power rpi
- yes
- (pin order is correct on the header and variant is satisfactory)

does killswitch kill uart when harness breaks?
- Yes, if the S pin gets unplugged from ground, the UART signal is killed. (didn’t test if some other pin falls out)
does pixhawk uart comms work when killswitch disabled via shorting header pins?
- the pixhawk and rpi comms should be validated on rev A boards (which are known good) to prove the rpi software is working correctly. We are trying to test the rev b board here and not the rpi software stack ya know …
- Yes, can read heartbeats and connected to the device, just no GPS data since in the bay
does pixhawk uart comms work with killswitch disabled via pixhawk gpio?
- Not tested, don’t know how to switch pixhawk gpio pins on/off
do power leds work
- Seems to work: green lights are always on, pi and pixhawk is powered on properly
  - Although sometimes the pi does say it does not have enough current (<5A), but that might be because the battery used was at 40%?
    - this was reported by autonomy people and seems like a major issue …. the rpi should be able to get all required current … EE is following up in chat
do uart indicator leds work
- Red LED is on when sending data from rpi to Pixhawk (working)
- Blue LED is typically always on
  - See comment in the schematic, this is due to an issue in the design that is not easily solvable