SingleCopter saturates servo outputs

ArduCopter Single/Coax
1

I am flying (sort of) a ducted SingleCopter running Copter 4.4.3. Here is what happens:

  • When unarmed, flaps respond properly to pitch and roll inputs, but not to yaw inputs.
  • When armed, flaps respond to yaw inputs also.
  • When armed in stabilized mode, the flaps begin to move to their limits in pitch and roll (but not yaw), over a few seconds.
  • When airborne the vehicle quickly stabilizes itself in yaw before abruptly pitching and rolling out of control (good thing I have a tough drone).

The silver lining is that I am no longer concerned about having sufficient control authority. I have seen this behavior on the bench and thought it might sort itself out in the air. Apparently not. So I seem to have messed up the configuration somehow. I attach parameters and a link to logs, one of which includes two “flights”.

A few notes on the vehicle. Torque is taken out by fixed stators upstream of the flaps. Since the flaps are relieved of that task they work symmetrically and linearly over their full deflection. The RCOUT logs show two flaps backing off slightly from max deflection, which is enough to trim the yaw. So the system seems to be working in yaw, but not in pitch and roll.

The FC is mounted in the duct at x= -10 cm, y and z = 0. It is oriented with the arrow facing rearward (-x) and down (+z), which I take to be AHRS_ORIENTATION 34. After calibration everything else works normally in Mission Planner with this setting.

This behavior seems sufficiently anomalous that I am hoping the answer will be obvious to somebody. Thanks in advance.

Esotec SingleCopter v1.param (19.0 KB)
https://transfer.pcloud.com/download.html?code=5Zon470ZHYOEVDOnJTQZB9iaZvwNIQOBIVLuFTVw8KvcgNu9O584X

2

Hello @esotec , welcome to the community,

Such copters are rare, and usually their main issue is that the flaps are too small. You said that the servos go to their limits, so saturation is happening.

But you did not post any photos, so all this is just guesswork.

3

Thanks for the suggestion, but the servos go to their limits with the vehicle sitting stationary on the ground. Flight experience so far suggests that the flaps are effective enough to nearly flip the vehicle right over. They just need to be properly controlled.

I will upload some pics soon, when I have some.