I’ve flashed the Kakute F7 with the “latest” 3.7 build online, ran through all the calibrations.
I mapped motor4 to servo 6 to make up for the fact that my front left M4 esc pad fell off and I rewired the ESC to M6 pad instead.
Given the diagrams online, showing motors A,B,C and D start front-right and then work around clockwise ending with D on the front left, everything tests fine. A/C spin CCW and B/D spin clockwise.
Its basically the same setup I had in betaflight.
However, it flips as soon as a give it enough throttle to hover. From what I can tell, the motors on the left side (C and D) just take off.
I’m running out of ideas. Does anyone have any other troubleshooting advice?
By chance have you mounted the FC in an orientation other than arrow forward? I don’t know if it’s analogous to Copter but I’m using a Kakute F7 AIO in a Plane and the motor output mapping was odd I believe because I have it rotated 90° (with the proper AHRS parameter set). After some frustration I used an O-scope to determine which M outputs corresponded with which Servox channel parameter. Got it working after that.
We also have it setup w/ arduplane and the output mapping was definitely off. I don’t recall exactly what the mapping was, but M1 != SERVO1, so we had to play with the servoX_function parameters as well. Our board was facing forward, but I don’t think that will have any effect on output mapping.
I think you are right that the orientation shouldn’t matter. For a V-tail I ended up with this vs what I expected (Mx)
M1-Right tail (M4)
M2-Aileron (M1)
M3-Left tail (M2)
M4-Throttle (M3)
I suppose the Hwdef configuration would have to be changed to revise this but it was a minor annoyance. I used a scope to map it but plugging a servo in to each output would accomplish the same thing.
I don’t have an oscope, but I have soldered on motors and escs so I can see which output goes where. In the motor test tool in mission planner, I’m able to test motors A,B,C and D and they indeed spin the motors in the right order. Could it be that this mapping is only when connected to mission planner?
My only remaining idea is to put it in a stabilized mode and rotate the quad with one of the arms down and try to figure out which motor it’s spinning faster to compensate.
That’s a strange one. I have a F4 Nano on a 130 size quad and the motor order had to be changed due to FC/4 in 1 ESC orientation but after configuring using motor test I worked this out. Can’t guess why it would work correctly in motor test yet not be right. Accel calibration good?