Almost lost my Spark over water, Compass and IMU errors, why?

[Pashasdad]

Was flying my Spark over the LI Sound this evening, the aircraft was about 340 ft high and 750 ft over the water from my location on the beach. I suddenly due to some unknown reason got a compass error followed by an IMU error and it switched into ATTI mode. I began to panic and barely got the aircraft back to shore safely after trying to fly in Sport mode to get back quickly.
Can't understand why I would get a compass error way out over water with no magnetic material anywhere close to the aircraft. This is why I suspect some latent defect. I have attached some possible useful photos below.

Could this be a defective aircraft? Can anyone help and/or recommend a course of action?

 

[Dronason]

You can upload the flight log from your app here DJI Flight Log Viewer and post the link to it. Then it can be checked more in details.

 

[Glen e]

This is the type of crap that really worries me, granted every forum reports problems with any drone, yet we are seeing this , in my Opin, way too often here....who the **** knows what "workingimu......blah, blah......" and dji says nothing......

 

[Pashasdad]

Here is a link to the flight data:

DJIFlightRecord_2017-08-08_[20-06-44].txt

BTW this was the second flight of that same day with a second battery. First flight was about 12 minutes long. There were no such issues on the first flight.

In fact these errors did not show up until after 11 minutes into flight #2. I would think if there was a compass cal problem it would have shown up much earlier. Calibration was done the day before in a wide open field, several flights were undertaken prior to this one, There was no need for compass cal just in case it is thought it was required.

In fact I had deliberately checked that the red directional arrow in the lower left box in DJI GO4 and it actually was pointing in the direction the aircraft was pointed to prior to launch! I always do this now after a prior issue with a fly away and crash on my original Spark that was replaced by DJI for this one.
This reduces the possibility of compass errors at calibration!

Additional flight data:
text file:
https://drive.google.com/file/d/ ... 2M/view?usp=sharing

Here is .dat file:
https://drive.google.com/file/d/ ... Ek/view?usp=sharing

Here is Black box Data:
https://drive.google.com/drive/f ... elA5QUk?usp=sharing

 

[Dronason]

GO4 flight record is has been uploaded HERE.

 

[Pashasdad]

Perhaps a forum member with some expertise in analysis can somehow investigate what this data tells us about the problem.
Thank you in advance !

 

[Dronason]

Will take some times to check but as you provided the .DAT file, there is usually way to understand what happened by some expert.

 

[motard]

Before flight, in ground ,you look lower left corner of the app to see if the pointer was pointing to the front off the aircraft? Or if you turn in hand the craft the pointer follow in the same direction? I hope you discover the problem to help us too. Regards.

 

[Pashasdad]

From a DJI Spark forum post:

In bottom left hand corner on your map you will see small red triangle, check to make sure that this triangle is pointing in the same direction (heading) as your AC, this will show good compass on the ground.

 

[Dronason]

Perhaps a forum member with some expertise in analysis can somehow investigate what this data tells us about the problem.
Thank you in advance !

It seems that your DAT file is too short for a so long flight, so it is either not the right file or it has been corrupted.
From the flight record uploaded, it indicates some Yaw error before there is the compass error that pop out. That's the usual error message in case of compass failure or defective calibration, as you were over water at some altitude.

 

[gnuPilot]

All of the videos on YouTube I’ve watched recommend calibrating the compass whenever you move to a new geographical location before flying there. Was the field you calibrated it at close to the beach you were at when event happened?

 

[Pashasdad]

Yes I was at the beach only 1 mile from home and did not recalibrate.

Seems that we are giving mixed messages at times. It has been stated that we don't need to recalibrate every time if not too far from usual flying location, possibly many miles away, maybe like 50 or so or more.
Then we have these types of events where it would benefit possibly to calibrate. Guess I'm confused!

 

[Glen e]

I agree, there are countless posts here that people have never calibrated the compass. When you take off and get an "ok to fly-GPS", that should be all you need. I always hover for 3 min or so at 15 feet to check everything, never take off with less than 12-14 sats.

 

[Dronason]

I agree, there are countless posts here that people have never calibrated the compass. When you take off and get an "ok to fly-GPS", that should be all you need. I always hover for 3 min or so at 15 feet to check everything, never take off with less than 12-14 sats.

If your compass is well calibrated, it will not change that easily. But the environment could be not that good and recalibration at that place could be the worst to do.
Hovering at some height, is always good to do, 3 minutes seems a lot as if you have an issue with the compass, it will be matter of seconds.
Another point is after lifting it up to verify that GPS home point is well recorded.

 

[Dronason]

Yes I was at the beach only 1 mile from home and did not recalibrate.

Seems that we are giving mixed messages at times. It has been stated that we don't need to recalibrate every time if not too far from usual flying location, possibly many miles away, maybe like 50 or so or more.
Then we have these types of events where it would benefit possibly to calibrate. Guess I'm confused!

The Dat file seems to be from the RC, not from the Spark. Nothing about your issue can be retrieved. How did you retrieved this Dat file?