Firmware version Pixhawk 1 2020-02-21 01.20
Hardware Pixhawk 2.4.8
Looks like EKF should be disabled by default
Glad to help developers in any way and thanks for your work,
Regards
aldo
1 Like
Hi Aldo,
Interesting find and video! I would like to see some photos of your design posted if you can.
What is the X8R receiver for?
Cheers!
Thanks for your interest.
X8R is to manually orientate pan-tilt from my Taranis X-Lite transmitter
Everything is designed using Autodesk Inventor
then is 3D printed in ABS
Pan motor is a http://www.micromotorssrl.com/motoriduttore_e192_ita.html with reduction 1:50 powered at 12V with encoder. The encoder’s feedback goes to a Pololu JRK 21v3 https://www.pololu.com/product/1392 that transform the motor to a continuous rotation servo
Then I have a 915 MHz Sik Radio on Telemetry2 to connect a Mission Planner to the Pixhawk relative to the Antenna Tracker frame. Another 433 MHz Sik Radio on Telemetry1 is dedicated for the connection with the airplane. A bluetooth module HC-05 permits a direct connection from Mission Planner to the 433 MHz Sik Radio mounted on the Antenna Tracker.
The Pixhawk of the Antenna Tracker has only an RX connection to the 433 MHz Sik Radio, TX is disconnected, so the Antenna Tracker con only receive infos from the airplane without any doubt.
RX and TX from 433 MHz Sik Radio are connected to an HC-05 that act as a COM port by-pass with my PC.
Copper grounded and alluminium are applied were oscilloscope with RF probe says that noise is present as 433 MHz are a damn source of noise
Tilt servo is a GWS S777CG 6BB Giant Servo https://www.pololu.com/product/516/specs
A quad directional antenna is on the way
I don’t think a bug, its just hard for the EKF if you spin it fast. This is not something it would expect to see in flight. You could try EKF3 it might be slightly better than EKF2.
The solution is probably just not to spin it fast for ages, that it not something it would try to do by itself.
Hi Greg, I’ve seen your works, much respect.
My Antenna Tracker is only a quick prototipe so the key concepts are low cost, low complications, continuous rotation pan, all in Autodesk Inventor, RF noise under control, low weight, no CNC milling, yes 3D print, no gears, no ball bearings, no stainless steel stuffs, nylon screws easy to cut.
A key concept (probably the only) is the continuous rotation pan junction 3D printed in ABS. Many hobbyists underestimate the ability of ABS to build extremely resistant and friction-free joints, as long as the speeds in play are not particularly high. Obviously keep the ABS away from the acetone … but at the same time this solvent offers the possibility to glue pieces in an exceptionally effective way using the same ABS.
This is the lowest ABS base:
The upper base that freely rotates over:
and the motor with an ABS 3D printed head attached to the upper base:
1 Like