Thrust asymmetry

Hi all,

Flying a Tarot 650 frame with 10Ah 4S, MT3510 700KV motors and 12" props, w/ a DYS gimbal, AUW about 3kg. Pixhawk, 7N GPS/mag on a mast, etc.

I had a (partly) failed motor which I've now replaced - in this case I was seeing lower thrust on motor 4 than 3, with higher thrust than expected on motors 1/2, as the graph shows. This manifest itself mostly as slow yawing and poor altitude control.

3691202491?profile=original

I chalked this up to an unhappy motor and have now replaced it. But I'm still seeing massive thrust asymmetry (ignore the flight mode warning)

3691202564?profile=originalI've attached two logs, 31.BIN is with the new motor and 24.BIN is with the old motor. The frame is pretty well balanced, weight-wise, and has no major prop obstructions on any of the four motors. I could imagine a bit of front/back asymmetrical thrust but as this is cross-axis it looks more like a software issue or other fault to me. The Pixhawk is running APM 3.2.1 with the EKF enabled; I flashed it and reset everything early on in my testing to ensure it wasn't something to do with my initial gain tuning or other config left over from the previous vehicle it had been mounted on.

Any ideas would be great to hear - I'm confused as heck at this point!

31.BIN

24.BIN

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • One possibility is if the motor mounts are not level. Some of the various Tarot mounts slip and if they are not level your thrust is going to be off.

    • This was my first thought - they're all levelled as precisely as I can relative to the main body of the frame, and the clamps are on pretty tight (with loctite on all of the bolts), so it's all secure. They don't appear to have budged a bit since I first assembled it.

      • i doesn't take much misalignment to see this, it is show classic symptom of a twisted motor.

        • How much are we talking? I've confirmed they're level to within half a degree of the main chassis plate, and visually they look perfect...

          • Another good way to check is to look down the end of each arm with the propeller tips pointing at a common point on each adjacent arm (such as the tip of the motor shaft/spinner) and look for a high side, spin the propeller to check its not a bad tracking prop.

            I really wouldn't be surprised if 0.5 deg was enough to get the difference you are seeing, especially if more than one arm is 0.5 deg and in the same direction.

This reply was deleted.

Activity