In-flight yaw alignment cause of crash?

This morning, after a successful “compass learn” flight to re-calibrate my compass, I started a second flight in loiter that quickly experienced loss of control.

As I’m a rookie at analyzing log files, I’d really appreciate some assistance at looking over my bin file from the flight.

I’m wondering if the “in-flight yaw alignment” that’s noted in the mavlink messages might have gone wrong. I had installed a new gps/compass mast this morning (Here3) and I’m wondering if perhaps the CAN connection to the Here3 might have come loose.

The bin file is uploaded to dropbox here: Dropbox - 2022-01-24 09-33-40 - crash - possible compass problem.bin - Simplify your life

And some screenshots that caught my attention are here:

RC8_OPTION = 31 (MotorEStop).
And you triggered the 8th channel on air.

Thanks for taking a look at my BIN file.

Yes - I selected Motor-Stop after losing positive control of the copter and it was heading for shrubbery. Unfortunately, I still damaged my carbon fiber props.

As you can see from the complete time frame - it was at the end of the flight - a very short flight.

Do you have any thoughts about the in-flight yaw alignment causing the loss of control?

Since the copter was in LOITER, it was as if it were beginning to toilet-bowl.

Thanks for any help or ideas you can offer!

Yes, I have some experience, in a very bad way.
We designed some quadcopter system without knowing that the power wires close the magnetometer mess up the magnetic field readings.
After takeoff, the vehicles were doing some rodeo moves.
Do you have any power wires close to the GPS module or autopilot?
If so, probably this is the cause, separate the autopilot and GPS module (the magnetometer) from power wires.

We were seeing that, too.

Thank you Mustafa -

This flight was in my back yard - no power lines. (unfortunately - just shrubs and trees!)

As I did the “compass learn” flight just before this one, I’m wondering if perhaps there’s a procedure I should have followed - perhaps I needed the copter to boot up wait a few minutes for any alignment to take place - and then launch.

One important lesson is that I probably should have just switched to STABILIZE mode - and then try to regain control.

I’m fairly new to learning how to analyze the log files. But I did notice on some of the MavExplorer graphs that the ATT.YAW data disappears a few seconds after the beginning of the time-line.

Again - thank you for taking the time to look into my log file!

By the way - I just took a look at your on-line software development course. It looks like something that would help me - I’m going to enroll.

1 Like

No, the power wires from battery to ESCs and also ESCs to motors.

Ah - no, I don’t think so.

All the power is handled on the bottom plate of this copter - the battery is below even that. I use MAUCH power sensor and BEC on the bottom plate. The ESC’s are on the bottom plate too.

The Orange Cube and radios (Crossfire and RDF900x) are mounted on the top plate. And I’d made several flights in this configuration.

There was one power related change a few weeks ago - I mounted a small 4S 420mAh battery on the top plate that powers the STorM32 gimbal. But it hasn’t caused problems for the 10 to 15 flights I’ve made with it installed.

The significant difference was the new mast I fabricated for the Here3 GPS/magnetometer. It’s made of carbon fiber tube, and it’s about an inch taller than the previous metal mast that was provided in this kit.

That’s why I did the compass learn on the previous flight - to recalibrate the compass in case it was in a different position a degree or two - and experienced less electrical interference due to the taller mast.

Because of moving the Here3, I’m wondering if perhaps the CAN connector wasn’t seated properly. I don’t know if the BIN file has any information about CAN activity - it would be interesting to find out.

Thank you!