Spark Power Failure During Flight

[cozzykim]

That's good to know.

What I'm suggesting though is not that the Spark turns on its belly, but that the software thinks that it has because of a bug or a glitch in the positioning sensors.

Just a thought, as I've been reading the reports with interest.

 

[Gabriel]

That's good to know.

What I'm suggesting though is not that the Spark turns on its belly, but that the software thinks that it has because of a bug or a glitch in the positioning sensors.

Just a thought, as I've been reading the reports with interest.

Nothing in the flight data indicated this as it would of showed up because the spark has a compass.,
This is why you calibrate it.

As for the remote, loosing video signal results in a RTH by default., NOT AUTOPILOT MODE which kicks in when the battery is too low because the aircraft has to land in time safely.

The reason why autopilot mode did not get recorded though is because the flight data gets disrupted when the remote and aircraft disconnect entirely which means the glitch in the firmware update caused this to randomly happen.

You can actually simulate this with your own spark if you have one., just shut off your remote while hovering and see if it goes into RTH mode ALL the time, someone just reported that the drone lost control the 4th time as if it had no gps!

This is where DJI is at fault, they sent out firmware updates which caused instability in the performance of their remotes, they literally ignored my questions about the remote alerts because it would be like admitting it was a malfunction caused by their own engineers, resulting in a recall of the aircraft and remote or a single case to replace mine at least.

 

[cozzykim]

Nothing in the flight data indicated this as it would of showed up because the spark has a compass.,
This is why you calibrate it.

Are we certain 100% that this is how the compass works, it's the IMU that measures orientation?

We also know that the IMU stops the motors if the Spark is upside down, that's what I'm suggesting, that it's possible that the IMU is confused and thinking that the aircraft is inverted.
Does the flight data include a record of orientation, I haven't really had time to look yet?

TBH, I sincerely hope that's not the case but, as these events seem to show that the craft has literally dropped from the sky, it has to be either a flight controller error or a power supply error turning the power off.

 

[Gabriel]

Are we certain 100% that this is how the compass works, it's the IMU that measures orientation?

We also know that the IMU stops the motors if the Spark is upside down, that's what I'm suggesting, that it's possible that the IMU is confused and thinking that the aircraft is inverted.
Does the flight data include a record of orientation, I haven't really had time to look yet?

TBH, I sincerely hope that's not the case but, as these events seem to show that the craft has literally dropped from the sky, it has to be either a flight controller error or a power supply error turning the power off.

All this is recorded in the flight data, yet the flight data is what was disrupted., flight data doe snot get disrupted because of app crash either, its stable, this is DJIs proof in the event anyone tries to fool their history, the only way the flight data would get disrupted is if the remote completely un synced itself from the aircraft., abandoned it., which is impossible for a user to do this during a flight...and we are not talking of the remote loosing battery life and shutting down which results in a RTH automatically, so the only logical explanation causing this is a malfunction in the remote or poor programing in the firmware.

 

[cozzykim]

Unlikely to be 'poor programming', more likely to be an unexpected combination of inputs and sensor routines.
That's what software or firmware bugs do, they can make the hardware behave in entirely unexpected ways.

I hope you don't think I'm a noob trying to argue or troll you, I'm trying to give you some support as this is obviously a very distressing problem for someone to lose an expensive piece of hardware without a satisfactory explanantion.

I'll leave it at that.

 

[Gabriel]

Unlikely to be 'poor programming', more likely to be an unexpected combination of inputs and sensor routines.
That's what software or firmware bugs do, they can make the hardware behave in entirely unexpected ways.

I hope you don't think I'm a noob trying to argue or troll you, I'm trying to give you some support as this is obviously a very distressing problem for someone to lose an expensive piece of hardware without a satisfactory explanantion.

I'll leave it at that.

i discussed starting a class action, i tried speaking to DJI about it but been the person on the phone insisted to relay messages to other over and over and over instead of putting me through to someone directly forget it!, this is reasonable enough to sue when you can not meet with dialogue.

 

[Vate]

I know this is an old thread, but a friend of mine just had the same happen to his Spark. As I also have a Spark, I thought I would research this problem. After reading this thread, it occured to me that one of my life's experiences may help explain an answer to this problem.
I have a full sized aeroplane that I built and it is powered with a Toyota Lexus V8 that is computer controlled. I spent over 4 months trying to track a problem where the motor would shut down for no reason and not throw up any fault codes. Fortunately I had a pretty primitive back up system that allowed an extended glide to get back to the airfield (in case you were wondering?). I replaced the computer three times... the last time to an after-market type, but still the same problem. Eventually I narrowed the problem to a relay, and on closer inspection found that the problem was in the mounting block to the relay. When you pushed the relay into the block, one of the female connectors clipped into the block would move down inside the block and so just touch the end of the terminal on the relay. Under certain conditions it would then lose contact cutting all power to the ECU shutting it down and as the system was wired in such a way, it also lost power to the memory of the ECU causing no fault codes to be logged. I fixed that problem, and after several years have never had a hint of a problem such as this since.
Hence this Spark problem to me sounds very much like a dirty battery contact allowing a loss of power long enough to shut the system down. Has anyone considered this?