Pretty much completed the Hexacopter build - DJI F550, E300 propulsion system, Pixhawk (fw version 3.1.5), Neo-7M GPS / Compass - but found something strange when bench testing. Motors were connected as per the hexacopter ‘X’ drawing / diagram at copter.ardupilot.com/wiki/motor-setup/ so the lead two motors when seen from behind are motors 3 & 5 with 5 being on the right front.
After arming and starting the motors - props attached - I proceeded to increment the throttle whilst operating the yaw / pitch / roll controls, I observed the following.
The ‘yaw’ increases motors quite violently, not convinced it’s doing the appropriate three motors though, even so I’m pretty sure this will yaw severely.
Elevator ‘stick up’ or ‘forwards’ depending on how you look at it the front pf the copter actually tried to pitch up - when I expected the rear to pitch up and front to pitch down - basically the reverse of what I expected. Yaw and roll seem to be OK.
I’ve reversed channel 2 (I use Mode 2 AETR) and the pitch is now as expected but my question is should I have had to do this - is there some error in my build? do I have motors 3, 5 & 4, 6 reversed - note I have connected motor 1 to Pixhawk output 1, motor 2 to 2 and so on in accordance with the previously mentioned document / link.
I’m concerned in case this build tries to spin up the wrong pair of motors i.e. if it needs to ‘lift the nose’ it will increase speed on the two rearmost motors - absolutely the opposite of what’s needed.
Well I’ve now flown it and it behaves OK so I think the reversal translates into the control software.
I’m tempted to try reversing the front and rear pair of motors then setting the channel to ‘normal’ forward control - I’m wondering if the Hexacopter orientation diagram is actually incorrect - i.e. motors 3 & 5 are actually the rear and 4 & 6 are the front.
The diagram is correct. It works with my Hex (on an APM).
Here is my suggestion:
1.Double check you do have the motors connected as in the diagram
2. Make sure your ESCs are calibrated
3. Make sure your motors are all spinning in the correct direction (as per the diagram)
4. Make sure you do all the PixHawk calibrations
If you are testing on the ground, it could be that the PixHawk is raising motor RPM to those levels because the copter isn’t able to YAW. I.e. it is trying to YAW, but its not happening, so it keeps raising the RPM of the appropriate motors.
100% certain connected as per that diagram which is why having to reverse a channel surprised me.
I just removed the top plate and re-verified. My motors are 100% as per the diagram based on motors 3 and 5 being the front. If they were the front I would not have needed to reverse channel 2 because my transmitter is not inverted for any channel.
DJI Opto ESC’s can NOT be calibrated, as for the motors and yaw the increase isn’t progressive enough for me to agree - they go to max almost immediately - well they did - but it was down to the update frequency of the ESC’s - the default on the Pixhawk seems to be 490 Hz - the DJI Optos don’t like more than 450 Hz.