Crash Investigation: Sudden Increase in Speed + Crash

Joined
Apr 22, 2019
Messages
8
Reaction score
0
Age
31
Hi Guys,

I've long read this forum, but first time posting...

I had a situation occur this morning which resulted in the death of my Spark. I'm typically a slow flyer, admitting I'm a relative beginner but something happened which I can't understand based on my Flight Logs. I'm hoping someone more experience

Here's what happened: I am currently on vacation in the highland regions of New Zealand.

I was flying my drone close to myself at my place of stay at a low altitude and distance to where I stood. The flight was no more than a minute or so, flown in ATTI mode (hovering and moving between 1mph to 4mph), when the drone dramatically sped up rising from 3mph to 19mph in the span of 1.3 seconds then again upon return 4mph to 16mph in 2.1 seconds - with the final acceleration resulting in a loss of control and crash.

I've investigated the flight records closely and cannot seem to understand the cause of the issue. I've flown the spark consistently through my trips with limited speed and did not expect it to accelerate uncontrollably at these speeds in such a sudden.

I've attached the flight logs for your reference as well. I did not recieve any environmental warning on strong winds to suggest otherwise - so I believe it might have been a defect or malfunction. According to my understanding of the drones features I do believe there should've been safety limits and consistent controls to ensue consistent flight behaviour.

Hope someone can help - I want to learn from this incident... I feel awful over the whole thing.
 

Attachments

  • DJIFlightRecord_2019-04-22_[09-38-23]log.zip
    65.9 KB · Views: 328
I was flying my drone close to myself at my place of stay at a low altitude and distance to where I stood. The flight was no more than a minute or so, flown in ATTI mode (hovering and moving between 1mph to 4mph), when the drone dramatically sped up rising from 3mph to 19mph in the span of 1.3 seconds then again upon return 4mph to 16mph in 2.1 seconds - with the final acceleration resulting in a loss of control and crash.

At ~74 seconds the aircraft experienced a Large Yaw Error ( Basically a compass error ). The aircraft then auto-switched to ATTI at ~80 seconds which means no GPS positioning. The aircraft was responding normally to your inputs and appeared to be under your full control. You may not have experience flying in ATTI mode. From what I see, the crash is a result of that and not due to the actual aircraft errors. The aircraft will drift with the wind when in ATTI mode. There was a brief period where the drift speed up occurred ( With no control inputs from you ) and this would be normal.

Capture.PNG
 
  • Like
Reactions: magica4
At ~74 seconds the aircraft experienced a Large Yaw Error ( Basically a compass error ). The aircraft then auto-switched to ATTI at ~80 seconds which means no GPS positioning. The aircraft was responding normally to your inputs and appeared to be under your full control. You may not have experience flying in ATTI mode. From what I see, the crash is a result of that and not due to the actual aircraft errors. The aircraft will drift with the wind when in ATTI mode. There was a brief period where the drift speed up occurred ( With no control inputs from you ) and this would be normal.

View attachment 110762
Amazing thank you for that explanation.

Do you reckon it would be worthwhile reporting it to DJI? It was still caused by an error, and the drastic increase in ATTI feels like a silly design?

I'm a little lost on the matter - I feel cheated and yet disappointed in myself on the matter
 
It was still caused by an error, and the drastic increase in ATRI feels like a silly design?
This depends on IF the error was due to your takeoff location. Usually these types of errors are caused by taking off ( Or calibrating the compass ) close to a mettallic object. Even re-bar in concrete can cause this. The aircraft is designed to auto-switch to ATTI mode due to the fact that it does not know direction. GPS positioning is disabled and it is up to the pilot to control the aircraft. It will maintain altitude via the barometer but positioning and direction are all up to the pilot.
 
  • Like
Reactions: magica4
This depends on IF the error was due to your takeoff location. Usually these types of errors are caused by taking off ( Or calibrating the compass ) close to a mettallic object. Even re-bar in concrete can cause this. The aircraft is designed to auto-switch to ATTI mode due to the fact that it does not know direction. GPS positioning is disabled and it is up to the pilot to control the aircraft. It will maintain altitude via the barometer but positioning and direction are all up to the pilot.
Well it was flying outside in central Queenstown. I dont think it took off near anything remotely abnormal. A lot of nature around here...

Perhaps a faulty GPS error?
 
At ~74 seconds the aircraft experienced a Large Yaw Error ( Basically a compass error ). The aircraft then auto-switched to ATTI at ~80 seconds which means no GPS positioning. The aircraft was responding normally to your inputs and appeared to be under your full control. You may not have experience flying in ATTI mode. From what I see, the crash is a result of that and not due to the actual aircraft errors. The aircraft will drift with the wind when in ATTI mode. There was a brief period where the drift speed up occurred ( With no control inputs from you ) and this would be normal.

View attachment 110762

The yaw error was irrelevant since the aircraft never had any GPS satellites - positioning was via the vision system and responded normally. Once the aircraft climbed that was lost, hence the switch to ATTI and the pilot lost control.

110765


The question is why did it have no GPS reception. Location, cold start or faulty GPS module/connector? The mobile device had a position lock:

110766
 
  • Like
Reactions: magica4
The yaw error was irrelevant since the aircraft never had any GPS satellites - positioning was via the vision system and responded normally. Once the aircraft climbed that was lost, hence the switch to ATTI and the pilot lost control.

View attachment 110765

The question is why did it have no GPS reception. Location, cold start or faulty GPS module/connector? The mobile device had a position lock:

View attachment 110766
Yes good question... I reckon it could be an issue with the GPS module? I'm dead centre near Queenstown, I dont think GPS would be a problem here. Further, this was all done outside - not sure if interference would've played a part, I wasnt under some concrete or anything magnetic...
 
Yes good question... I reckon it could be an issue with the GPS module? I'm dead centre near Queenstown, I dont think GPS would be a problem here. Further, this was all done outside - not sure if interference would've played a part, I wasnt under some concrete or anything magnetic...

Ignore the compass error - that happened before the aircraft took off and could have been for any number of reasons. Why did you take off without any GPS satellites locked? Has that happened before? Will the aircraft still power up and, if so, does it show any GPS satellite reception?
 
The yaw error was irrelevant since the aircraft never had any GPS satellites - positioning was via the vision system and responded normally. Once the aircraft climbed that was lost, hence the switch to ATTI and the pilot lost control.

View attachment 110765

The question is why did it have no GPS reception. Location, cold start or faulty GPS module/connector? The mobile device had a position lock:

View attachment 110766
Do you reckon this is something DJI can assist me on?

I've been looking up online and so many people have had similar issues... but I'm worried how helpful DJI could be on the matter.

To me it felt like a normal and short flight in a small distance. I wasnt doing anything dangerous in odd areas - but the suddeb switch to ATTI and loss of control due to lack of GPS would have been major contributing factors...

I used to love flying drones, but now I dont know if I can trust DJI technology...
 
Do you reckon this is something DJI can assist me on?

I've been looking up online and so many people have had similar issues... but I'm worried how helpful DJI could be on the matter.

To me it felt like a normal and short flight in a small distance. I wasnt doing anything dangerous in odd areas - but the suddeb switch to ATTI and loss of control due to lack of GPS would have been major contributing factors...

I used to love flying drones, but now I dont know if I can trust DJI technology...

Unfortunately the problem was fundamentally an error on your part - you took off with no satellites locked. None. That guarantees ATTI mode once you lose VPS, and if you are not experienced flying in ATTI mode it will most likely result in a crash.
 
  • Like
Reactions: magica4
Ignore the compass error - that happened before the aircraft took off and could have been for any number of reasons. Why did you take off without any GPS satellites locked? Has that happened before? Will the aircraft still power up and, if so, does it show any GPS satellite reception?

Think it has happened before, it goes on and off but I've always been able to maintain control and return it - but I've never had it result in such a tremendous acceleration
 
Unfortunately the problem was fundamentally an error on your part - you took off with no satellites locked. None. That guarantees ATTI mode once you lose VPS, and if you are not experienced flying in ATTI mode it will most likely result in a crash.
I see, well thank you for your insight.

As a relatively new flyer this was a tough lesson to learn...

But I'm glad I at least have an answer it seems.
 
Think it has happened before, it goes on and off but I've always been able to maintain control and return it - but I've never had it result in such a tremendous acceleration

The acceleration was caused by your elevator input once the aircraft switched to ATTI. Unfortunately the aircraft is well out of warranty, so even if you can demonstrate and persuade DJI that the problem stemmed from a faulty GPS module they are not going to replace it.
 
@sar104 Hmmm....When l look at this with the .txt log tool. This is different than the standard .csv file. Is this unique to the Spark, coding wise for CSV View? According to the .txt tool decoding, there were 13 sats aquired at takeoff and the errors did not occur until well in to the flight?
Capture2.PNG
 
@sar104.....My bad. I found the reason why. The data I loaded was from another spark flight. My appologies to @magica4 as well.
 
If you took off between high objects to block satellites, you will not have GPS.

That's not what happened, as illustrated by the fact that the mobile device on the ground had good satellite reception. Zero satellites is rare, and suggests that something else was going on.
 
  • Like
Reactions: Quest-So
Hi Guys,

I've long read this forum, but first time posting...

I had a situation occur this morning which resulted in the death of my Spark. I'm typically a slow flyer, admitting I'm a relative beginner but something happened which I can't understand based on my Flight Logs. I'm hoping someone more experience

Here's what happened: I am currently on vacation in the highland regions of New Zealand.

I was flying my drone close to myself at my place of stay at a low altitude and distance to where I stood. The flight was no more than a minute or so, flown in ATTI mode (hovering and moving between 1mph to 4mph), when the drone dramatically sped up rising from 3mph to 19mph in the span of 1.3 seconds then again upon return 4mph to 16mph in 2.1 seconds - with the final acceleration resulting in a loss of control and crash.

I've investigated the flight records closely and cannot seem to understand the cause of the issue. I've flown the spark consistently through my trips with limited speed and did not expect it to accelerate uncontrollably at these speeds in such a sudden.

I've attached the flight logs for your reference as well. I did not recieve any environmental warning on strong winds to suggest otherwise - so I believe it might have been a defect or malfunction. According to my understanding of the drones features I do believe there should've been safety limits and consistent controls to ensue consistent flight behaviour.

Hope someone can help - I want to learn from this incident... I feel awful over the whole thing.
Sorry to hear about the loss of your Spark. However perhaps others can use it as a learning experience. There are certain things I drill into people when giving lessons. Among them are:
  • Never take off until you have GPS lock.
  • Never take off until your home point has been recorded.
  • Never take off with anything less than a fresh fully charged battery.
  • Never fly your drone into a position that puts a large obstruction between it and yourself.
  • Always check your RTH settings prior to every flight.
  • Always check your sensors (compass, gyro, accelerometer) prior to every flight. Don't take off unless all are GOOD.
  • Never start up the drone with the gimbal lock still on.
  • Never use cheap props or third party batteries (my personal point of view).
Those top 4 have caused many people to lose their drones. Taking off without GPS lock puts you immediately into a forced ATTI mode. This can and does cause issues for the inexperienced pilot. The issue is further compounded by the fact that with the Spark there is no simple method (i.e. switch on controller like on the Phantom) to manually practice flying in ATTI mode.

I realize that the Spark is designed to be simple and aimed at beginners. As such I can see the point of view from DJI being that a beginner doesn't need ATTI mode. However I look at it on the opposite side, in that a beginner needs ATTI mode in order to practice when GPS is indeed available. That way you can practice in ATTI mode and quickly switch back to GPS mode if/when you get in trouble.
 
  • Like
Reactions: Alvaro L
  • Never take off until you have GPS lock.
  • Never take off until your home point has been recorded.
  • Never take off with anything less than a fresh fully charged battery.
  • Never fly your drone into a position that puts a large obstruction between it and yourself.
  • Always check your RTH settings prior to every flight.
  • Always check your sensors (compass, gyro, accelerometer) prior to every flight. Don't take off unless all are GOOD.
  • Never start up the drone with the gimbal lock still on.
  • Never use cheap props or third party batteries (my personal point of view).
Nice List. Concise but covers most of the failures I read about up in here
 
  • Like
Reactions: David_Cambridge
Well it was flying outside in central Queenstown. I dont think it took off near anything remotely abnormal. A lot of nature around here...

Perhaps a faulty GPS error?
Yes good question... I reckon it could be an issue with the GPS module? I'm dead centre near Queenstown, I dont think GPS would be a problem here. Further, this was all done outside - not sure if interference would've played a part, I wasnt under some concrete or anything magnetic...
If the location you showed was your takeoff point, it shows a lot of non-nature around.
Being under concrete is not what was meant.
Launching from on top of reinforced concrete is a very common cause of the the issue you experienced.
Or anywhere else that's close enough to something steel that can disrupt the compass.

The speed you mentioned matches your right joystick inputs and just shows the drone continuing without being able to stop and hold position.
Once your drone is in atti mode, it does not have the benefit of position holding that GPS gives and flying is like driving on ice with no brakes.
The drone can't stop and will continue coasting after you centre the joysticks.

The GPS is usually very reliable and the issue is much more likely to have been caused by where you launched than a faulty equipment.
 

Recent Posts

Members online

No members online now.

Forum statistics

Threads
143,066
Messages
1,467,352
Members
104,933
Latest member
mactechnic