Welcome DJI Spark Pilot!
Jump in and join our free Spark community today!
Sign up

Spark gone rogue...fly away

DJSparxx

Member
Join
Aug 25, 2019
Messages
9
Welp, my spark led a relatively long, fruitful life of 4.5 years before it went rogue and flew away from me.
Here's the scenario.
Was flying for a few minutes 100' in the air at a location I've flown hundreds of times. Battery was at 75%.
Spark suddenly goes in ATTI mode and starts jetting as in Sport Mode towards the north. BTW, I never fly in Sport Mode, so this spooked me. I lose sight as it goes over a small building. Luckily, someone points out it landed on a roof. The camera is still working as I can see the asphalt roof tiles on the GO4 app. When I retrieved it from the rooftop the following morning, I find no damage to the exterior. It looks like it landed softly on the roof as if it was RTH. BTW, I never RTH...always land manually. Also, before I launch, I always set the RTH every single time in case something crazy like this happens. Well that was a big fail!
Took it to my local drone shop for root cause / repair estimate. They couldn't explain what happened since it was sent directly to DJI for repair estimate. Repair Estimate involves replacement of the R Spark Core Board, nothing else.
I can live with the 4.5 year life cycle, but more curious why this incident happened.
 
Had about same experience with atti.
I guess it had to do with getting it too soon in the air without waiting for enough sattelites. Went to,the north too with the speed of wind.
if in atti mode the spark follows the wind which comes usually from the south west in Belgium.
About the landing itself.
I learned especially in winter time, never start a flight with a battery which isn.t charged 100%
During the take off and the warming up period of the ( smart) battery. There is very temporaly a high current drain. Dji thinks battery is dying and automatically sets in landing procedure. If you are still connected , one can manually force rising up again
it.s not typically spark but even mavic and phantom Have this problem
I have the impression in litchi this current drain is ignored.
open your flight data and see if there was a current drop in your battery
 
  • Like
Reactions: Spark 317
My spark is still at the drone shop for me to pick it up, so don't have access to flight log. I launched with the battery at 100%. Battery was at 75% when it flew away from me.
 
My spark is still at the drone shop for me to pick it up, so don't have access to flight log. I launched with the battery at 100%. Battery was at 75% when it flew away from me.

Follow the instructions here and retrieve the flight logs from the device you flew with. You don't need the Spark.


If you'd like for us to take a look, copy and paste a link back to here.

Good luck. :)
 
Hi, DJSparxx.

I'm no expert at this, so others may have a better read on the files than I do, but I'm interested in being able to understand these so i thought I'd have a look.

If I'm seeing this right, the log file shows takeoff 82 seconds after power-on, ascending up to around 45' above take-off. Then, at 102 seconds, it accelerates from 6mph to 20 in 0.12 seconds. That's over 5g in acceleration, which seems pretty fast to me - way faster than the wind speed could cause. Again at 105 seconds, another 5g+ acceleration, followed by a smaller (but still rapid) deceleration at 106 seconds. This might be hitting a power line on the back of Durfee Ave just before it crosses Walnut avenue.

The battery was still at 98% when the flyway occurred, and at 91% when the a/c finally came to rest on the roof.

GPS was good, with 18 satellites, and the home point was set. It looks like the home point was on some old railway tracks - does that amount of metal not interfere with the compass?

Screenshot_2021-01-13 Airdata UAV - Flight Data Analysis for Drones.png

The .txt log file may have more data(?); I've passed this one through AirData to convert to CSV and then uploaded that to the Phantomhelp.com viewer linked above.
 
My 2 cents,
as the spark crossed the railroad and flew over the parking , there was a massive current drain and spark descended from about 12 meters to 4 meters. ( DJI fail mode ?) Then indeed another current drain ( or voltage drop) as speed was going up at an altitude of 4meters it probably hit or a tree or as suggested a power line;
total signal drop at landing place ( solar cells ??) Wind came from south so without positioning it's flying to the north.
Strange no errors are seen
 
anyway , as I streetview the neighborhoud its all wires and powerlines . I would never fly in such a place . I think spark hit something while flying at 4 meters or just before which caused the "landing"
 
I manually calibrate the compass before every launch, and never had any issues before at this location with compass interference until that one time. The Spark was on even ground on launch...the google map view that was posted is old. Tracks are not there anymore.
BTW, I retired old Sparky and got myself a MM2.
 
  • Like
Reactions: Spark 317
Hi, DJSparxx.

I'm no expert at this, so others may have a better read on the files than I do, but I'm interested in being able to understand these so i thought I'd have a look.
My 2 cents,
as the spark crossed the railroad and flew over the parking , there was a massive current drain and spark descended from about 12 meters to 4 meters. ( DJI fail mode ?)
Hey Nik & Roland,

If you're into reading log files etc. you may find the following few links interesting and helpful:

Mavic Flight Log Retrieval and Analysis Guide
FOUND! Lost my Mavic Pro during storm
CsvView and DatCon

I've been meaning to put some more headtime into this but I always get diverted by something else.
 
Welp, my spark led a relatively long, fruitful life of 4.5 years before it went rogue and flew away from me.
Here's the scenario.
Was flying for a few minutes 100' in the air at a location I've flown hundreds of times. Battery was at 75%.
Spark suddenly goes in ATTI mode and starts jetting as in Sport Mode towards the north. BTW, I never fly in Sport Mode, so this spooked me. I lose sight as it goes over a small building. Luckily, someone points out it landed on a roof. The camera is still working as I can see the asphalt roof tiles on the GO4 app. When I retrieved it from the rooftop the following morning, I find no damage to the exterior. It looks like it landed softly on the roof as if it was RTH. BTW, I never RTH...always land manually. Also, before I launch, I always set the RTH every single time in case something crazy like this happens. Well that was a big fail!
Took it to my local drone shop for root cause / repair estimate. They couldn't explain what happened since it was sent directly to DJI for repair estimate. Repair Estimate involves replacement of the R Spark Core Board, nothing else.
I can live with the 4.5 year life cycle, but more curious why this incident happened.
You had better luck than I did. My Spark did the same less than an hour out of the box. I didn't use RTH either. Mine not only flew away, it reset its home point deep in the forest where it tried to go. A big red cedar got in the way, and it roosted there for months before a storm blew it down.
 
Baiklah, percikan saya menjalani kehidupan yang relatif panjang dan bermanfaat selama 4,5 tahun sebelum menjadi nakal dan terbang menjauh dari saya.
Berikut skenarionya.
Terbang selama beberapa menit 100 'di udara di lokasi yang telah saya terbangi ratusan kali. Baterai berada di 75%.
Spark tiba-tiba masuk dalam mode ATTI dan mulai melayang seperti dalam Mode Olahraga ke arah utara. BTW, saya tidak pernah terbang dalam Mode Olahraga, jadi ini membuat saya takut. Saya kehilangan pandangan saat melewati sebuah bangunan kecil. Untungnya, seseorang menunjukkannya mendarat di atap. Kamera masih berfungsi karena saya dapat melihat genteng aspal di aplikasi GO4. Ketika saya mengambilnya dari atap keesokan paginya, saya tidak menemukan kerusakan pada bagian luarnya. Sepertinya itu mendarat dengan lembut di atap seolah-olah itu RTH. BTW, saya tidak pernah RTH ... selalu mendarat secara manual. Selain itu, sebelum saya meluncurkan, saya selalu menetapkan RTH setiap saat jika terjadi sesuatu yang gila seperti ini. Nah itu kegagalan besar!
Membawanya ke toko drone lokal saya untuk perkiraan akar masalah / perbaikan. Mereka tidak bisa menjelaskan apa yang terjadi karena dikirim langsung ke DJI untuk estimasi perbaikan. Perkiraan Perbaikan melibatkan penggantian Papan R Spark Core, tidak ada yang lain.
Saya bisa hidup dengan siklus hidup 4,5 tahun, tapi lebih penasaran kenapa kejadian ini bisa terjadi.
SSpark suddenly goes into mode meaning that the GPS signal is disturbed or can't get a GPS signal, the cause may be thick clouds, electromagnetic GEO, electromagnetic radiation so it is difficult to get GPS
 

Members online

No members online now.

Forum statistics

Threads
14,593
Messages
118,799
Members
17,987
Latest member
csdisme