I recently converted a DJI S1000 for use with the Pixhawk system. I am having issues controlling the copter’s yaw in Stabilize mode. I have reversed the direction of each motor in order to work with the pixhawk’s expected motor direction. In some tests I can get the copter to stay pointed straight but the yaw stick is nearly fully deflected and I have seen motors almost completely stop spinning in order to counteract some torque. I have tested each of the motors and they are working properly. I have an IR camera triggering device and an RTK GPS system onboard. I’ve had a fair amount of experience with the pixhawk in various fixed wing and multirotor platforms but I can’t figure this one out. I do not believe it is a hardware issue. Any advise would be greatly appreciated. I can post video and logs if it would help.
It seems like this would have to be related to the accelerometer? Maybe I should reflash the firmware and start fresh?
Here is the log from the last flight before breaking the landing gear. There are several short hops in this log. the last one displays the most aggressive yaw.
I wonder if it’s possible the S1000 is designed with motor tilt to assist in yaw control, and by reversing the motors, you have created a situation where the motor torque is being countered by the motor tilt/twist?
I just looked at the log, and I think you have Quadcopter code loaded?
First clue, was that motor outputs 5, 6, 7 and 8 are always zero. Then I looked at the Frame Type definition in the startup message, and it says QUAD so…
Seems odd though, I’m surprised it would fly at all, or that you wouldn’t have noticed 4 of the motors aren’t turning?
Interesting… All 8 motors spin. Maybe only 4 are accelerating past the idle point after arming? The motor tilt could be another issue. I suppose I would have to reverse the motor direction again then swap each arm. I will also reinstall the firmware and see where that gets me.
In this log we can see the motors being spun up numerous times but it doesn’t look like it’s getting off the ground. The RATE message shows the yaw control is backwards as well though so I guess the pilot is spinning up the motors and finding that it’s rotating and puts it back down again.
It really must be a problem with the motors order, the spin direction or the propeller mounting. copter.ardupilot.com/wiki/connec … nd-motors/
Also make sure your vehicle is a “+” or “x”. The config shows FRAME = 1 which is correct for an “x” frame. I think an s1000 is an “x” but maybe when you were doing the motor order you were looking at the “+” setup? I’m not sure of course, I’m sure you’re being careful but it’s just got to be the motor order, the motor spin direction or the prop mounting that’s the problem.
Be sure you have the motor rotation as well as number correct. The s1000 frame numbers and esc signal connector numbers are not the same as on Pixhawk.
Performing a motor test in Mission planner shows the “A” motor in the front right position. Going down the list goes around the frame in a counter-clockwise direction. I’m confident that the motor directions and propeller rotations are correct. The problem must lie in the torque of the motor fighting the horizontal component of thrust from the motor tilt. I have un-reversed all the motors, swapped propellers, and swapped motor arms around to correct the counteracting torque/thrust issue. I also noticed there is a new firmware update, so after I update the firmware (if the weather ever clears up) I will give it another test flight.
I have built about a dozen S1000s up with Pixhawk and haven’t had any troubles, but I never swapped ESC wires or props just whole arms. The newer S1000 arms don’t even have any sideways angle. Have you had a chance to fly it again?
Also, other than setting up the Pixhawk wrong, a main reason for Yaw problems is miss-aligned motors.
Please check that your motors are level parallel to the center plate (I think that frame had dehedrial and therefore it won’t be level perpendicular to the center plate).