Having some bizarre issues with my quadcopter - mostly manifesting in stage yaw behavior.
Recently did a complete rebuild with a new frame, and now when yawing, to the left, the copter wobbles and overshoots the heading by as much as 180 degrees. Yaw to the right doesn't seem to have the same issue, but rotational rate is very slow initially, speeding up to match left yaw rate after a few seconds.
On landing, I noticed that front right and rear left motors were slightly warm, while front left and rear right were cold.
Looking at the logs, the rcout of all motors are wildly varying in their averages. I would have assumed they should be similar (like they are on my hex)
System has been wiped and had all re calibrations done - including esc, accel/gyro etc.
Thought it might be a mechanical issue and looked at the motors (which are new) but can't find anything there. System was a little nose heavy so I added some weight to the back and took it out again.
This time, the motors clearly run in pairs, with rcout ch1&2 running pretty close together, and 3&4 the same as well, but there is a massive difference in the outputs
Further, in this flight the copter randomly started to yaw gradually to the left. I looked at the yaw and desired yaw vs yaw input, and while the yaw follows the desired yaw almost perfectly, this doesn't match the yaw input. This has happened in previous flights as well. Given that the copter was in alt hold for this, I can't work out why the desired yaw would be moving on it's own.
Pretty much out of ideas as to what the problem could be here. Very frustrating, as the copter seems to fly pretty well mostly, except for yaw. I've tried autotune, and it just made a mess of it. It seemed to be pitching slightly when doing the roll tuning as well so I think there is a problem at a lower level than that.
Build:
Frame: Flyker Scorpion X4
Motors: Multistar Elite 4006-740KV
Props: 12x4 Carbon
ESC: T-Motor S25A
AP: Pixhawk AC3.2
GPS: 3DR GPS/Compass
Battery: Multistar 6600mAh 4S
I've uploaded the flight logs from the two flights described above. There is a crash at the end of the second flight which was entirely my fault. Didn't check the battery voltage properly, and the copter when into RTL failsafe under a tree.
Any advice would be much appreciated.
Replies
I've since tried replacing a motor and esc (motor 1) to try and eliminate any potential issue with motor mechanics, but the problem persists. It has to be a software issue, but I can't see where it could be coming from.
Anyone else have any ideas?
It looks to me like you have not balanced your quadcopter, so the weight is not centered between all of the motors. This would cause the two motors to have to work much harder to keep the aircraft level (hence channels 1 and 2 running at almost at max). Adjust the battery position until you have all four motors run at similar levels during hoover. The balance doesn't have to be perfect, but it really needs to better that what you currently have. This should help your yaw problem and also solve the problem with pitching during auto tune.
That's where I started. I had mostly assumed this was a mechanical issue, however:
1. It doesn't explain the desired yaw drifting on its own
2. For the second flight at least, the balance point is in the exact center of the 4 motors.
3. If I understand the logs correctly, motors 1&2 would be rotating in the same direction - indicating an imbalance in yaw output. How do you even create a physical imbalance which would manifest in a persistent yaw?
You're right about 1 and 2 spinning the same direction. I should have checked the motor number diagram as I incorrectly assumed they were adjacent motors and so would have spun opposite. So having the weight off center does not explain the issue you observed.