After an epic hack session last night, in the wee hours of the morning
we successfully demonstrated a running control loop on the new flight
The full data path of all critical systems has now been shown to work
in theory. It remains to be decided what exactly will fly depending on
how much testing we can get done in the next couple of days.
## Summary of Last Night's Test
After some wrangling and fixing an (unrelated to roll control) issue
with reading the GPS over USB we compiled Dan's roll control library
and did all the 'wiring' in the Flight Computer Framework (FCF) to get
IMU data from the ADIS to the RC and then to the servo board.
We spent some time testing with an oscilloscope looking at the PWM
until it looked like it was doing something reasonable. Then we hooked
up the actual RC servo, held the launch detect tether closed and ran
the FC. Sure enough the servos woke up and the fins did their
After opening launch detect the fins twitched and responded to
rotating the avionics module, clearly trying to keep the thing
straight. After a few seconds the fins clocked hard to the side in an
attempt to make the first pre-programmed roll. Because of wiring we
couldn't really test spinning the rocket. After 30 seconds the fins
went straight again and the program timed out (for apogee)
At the same time I ran the simple python packet reader "on the ground"
and saw valid packets getting unpacked. This is written very similarly
to the CS capstone's code so I fully expect their demo visualizer to
"just work" once we get it installed.
So tonight we saw the full path
IMU -> STM32 -> Ethernet -> FC -> Framwork -> Roll Library -> Ethernet
-> STM32 -> Servo
work, best we could tell, as expected.
In addition all the data was being logged to a logfile on the FC and
the Rasberry Pi on the Track Master received the same data over the
## Next Steps:
We still need to clean the code up considerably before launch. In
addition we would love some pretty intensive testing over the next 3
days. Anyone that wants to work on testing the framework should help
The avionics are getting a final check up and being bolted down and
buttoned up for flight tonight at the PSBA. This means we should be
showing up on launch day with a more or less finished rocket. Besides
the usual things that always seem to go wrong we're actually pretty
well prepared for this flight.
Great work everyone!! Keep it up and see you over the next couple of days.
psas-team mailing list
This list's membership is automatically generated from the memberships of the
psas-airframe, psas-avionics, and psas-general mail lists. Visit
http://lists.psas.pdx.edu to individually subscribe/unsubscribe yourself from