mintyfresh Posted at 2017-12-8 07:28
Upgraded to latest firmware. My drone also fell from midair from about 52m onto a tin roof. Battery and drone found separately. Battery on when collected. Case no. 873746. First night of flying. Thankfully recovered next day and surprisingly working. Very disappointed.

Here is the Google Drive folder showing all downloaded records of the crash

https://drive.google.com/open?id ... wGiawN6hpPUEEtU_ru-

Matioupi Posted at 2017-7-30 21:12
About all those falling Spark cases, it seems that most of time, they were just hovering (at least many people say so). Is this feeling right or were some of the machines moving when the shutdown occured ? Were they really hovering without stick or other means inputs or were they getting up or down ?

Mine was climbing and move away from me when it dropped

Went from auto track to fun mode (sport) and after about 15 sec it just dropped , of course I was flying over water at the moment...I live in Sweden so it was really  refreshing to jump in and save it.....gonna let it dry and try to start it in a few h ,but I can’t trust it after this. Maybe next time it fall and someone get hurt does DJI take responsibility if that happens/no they don’t... think I return it and get another brand otherwise it’s a great machine

My drone dropped from greater than  120m into dense bush, I have spent 3hrs bush bashing to the spot where the drone dropped but the bush was too dense to find the drone.

I can see from the logs on the phone that it falls after I accelerate back to my home point, battery was greater then 30%.

DJI have said that since I was in sports mode I must have crashed into something  because the front sensors were not operating! This  is a little absurd as I was at least 120m above ground it fell over 200m before losing the signal (over a cliff into bush). The logs seem to clearly show it dropping out of the sky!

Dji has offered me a 15%! For something that clearly on the logs seems to be an issue with their  drone. Absurd, I would not recommend buying this drone if this is the sort of warrenty  cover that you get!

Dead drop from 150 ft today, shattering my camera housing.  

I launched my spark today and it immediately started rising without any RC input.  It climbed to about 150ft and then simply started plummeting from the sky, crashing into a roof, smashing the camera housing.  Video was recording the entire time and battery was fully charged.  

This is really disconcerting.  If the drone hit someone on the head, it could kill them, easily.  If it hit a car moving at 60 mph, it could create a multi-car accident.  This is a major liability that Spark needs to address.

Armaneo Posted at 2018-1-1 11:58
Dead drop from 150 ft today, shattering my camera housing.  

I launched my spark today and it immediately started rising without any RC input.  It climbed to about 150ft and then simply started plummeting from the sky, crashing into a roof, smashing the camera housing.  Video was recording the entire time and battery was fully charged.  

i found it odd to design a drone that you control with your bare hand just a couple of feet away.
with palm launch and land, i'm surprised there hasn't been any law suits for injures.

I thought the errors with the drones suddenly dropping down the skies was fixed, but now I have flown my drone (spark) the last 3 days in a park with 2 12 minutes batteries and atleast one time a day it has suddenly just dropped down the sky ???? How in the world can that happen ?

This is definitely a case for not flying out of line of sight. Don't trust the technology fully, or there will be bad consequences

This is definitely a case for not flying out of line of sight. Don't trust the technology fully, or there will be bad consequences. Hopefully this issue is resolved because I dont want my SPark falling.

Same here   Dead drop from 30 m. The battery was 95%, fortunately it happened front of my house and nobody was there. Suddenly the propeller stopped and dropped down from sky. Landed on the top. Popped in the battery but the fan had horrible noise, but worked. I contacted with local (Hungarian) distributor company and they sent back. I have checked my case but nothing since 22/01. I hope it will solve very soon....    But it is very frustrating.  It could be happened above a water or any dangerous place.... Someone could have been hurt ...
Edited: 05/02/2018. DJI finisehed the repair process for free of charge. I'm waiting for the package.

Edited 2: Yesterday I received a brand new Spark. Thank you DJI!

Hi everybody!!
Here another case of Spark falling off after motor (noise) suddenly stopped when it had just started flying a minute or two, it was doing 75 meters up and all the sudden the noise stopped and we looked at it terrified to see it falling in front of us in the water, 3 meters from us. I immediately got it but it was too late, already damage.
What can I do??? How can I prove what happen??

fans26ecc6ed Posted at 2018-2-21 14:54
Hi everybody!!
Here another case of Spark falling off after motor (noise) suddenly stopped when it had just started flying a minute or two, it was doing 75 meters up and all the sudden the noise stopped and we looked at it terrified to see it falling in front of us in the water, 3 meters from us. I immediately got it but it was too late, already damage.
What can I do??? How can I prove what happen??

Uploading your logs to the following site and then paste the URL back here on this page.  Lots of people on here enjoy looking at the logs to see if they can be of help.

http://www.phantomhelp.com/LogViewer/Upload/

As you can see by these records I wasn’t very high or a long distance away, when my DJI spark just fell out of the sky and smashed the motor arm. I am wondering why this is the case as I’m new to flying and don’t want this to happen again it’s only a new drone and I paid a lot of money for this, where can I get this repaired as I’m in Christchurch New Zealand, or is this under warranty due to not my fault.

Dead drop from 43m today, without errors or warnings, just like rock.

Here my Flight log - http://www.phantomhelp.com/LogViewer/6RZV2K7G12SK1AR7MZGZ/

Unfortunately the locality is swamped and I can not find it

Tomorrow, in the light, I will try to look for it again. How it works in this case, I'm sure this is a dron problem, not my fault.

They don’t update outcomes in the forum....

Just for the record Spark owner since December 86 flights and not an issue, well that wasn’t pilot error...

Drones need respect and patience

georgio_iv Posted at 2018-5-1 12:46
Dead drop from 43m today, without errors or warnings, just like rock.

Here my Flight log - http://www.phantomhelp.com/LogViewer/6RZV2K7G12SK1AR7MZGZ/

Open a case on this one.  Your battery stopped reporting remaining capacity after it hit 82% and my guess is that it died in flight and the app didn't know it was that low and RTH wasn't prompted or initiated.  This was probably noticeable on your screen along with the longer than normal flight time but guessing you missed it while trying to fly/record/etc.

At least you proved that the Spark really does have a 15 minute flight time!

Armaneo Posted at 2018-1-1 11:58
Dead drop from 150 ft today, shattering my camera housing.  

I launched my spark today and it immediately started rising without any RC input.  It climbed to about 150ft and then simply started plummeting from the sky, crashing into a roof, smashing the camera housing.  Video was recording the entire time and battery was fully charged.  

SPARK drfting without an RC input and picking speed may be a case of connected internet. Disconnect the internet while flying.

Hylle Posted at 2018-1-17 12:44
I thought the errors with the drones suddenly dropping down the skies was fixed, but now I have flown my drone (spark) the last 3 days in a park with 2 12 minutes batteries and atleast one time a day it has suddenly just dropped down the sky ???? How in the world can that happen ?

disconnect internet please.

My brand new Spark just fell out of the sky on it's first flight.  I flew it a short distance - no interfering signals, no metal anything, no trees above, nothing - and it just dropped out of the sky and shut off.  I was using the controller and both it and the drone had plenty of charge.  Blackbox data submitted to DJI a few moments ago.

http://www.phantomhelp.com/LogViewer/KWLSGPTEF8FIFFE5042W/

k9education Posted at 2018-5-4 00:28
My brand new Spark just fell out of the sky on it's first flight.  I flew it a short distance - no interfering signals, no metal anything, no trees above, nothing - and it just dropped out of the sky and shut off.  I was using the controller and both it and the drone had plenty of charge.  Blackbox data submitted to DJI a few moments ago.

http://www.phantomhelp.com/LogViewer/KWLSGPTEF8FIFFE5042W/

Contacted DJI and they would not even look at the logs.  They wanted me to send it in at my own expense (less than 24 hours after I received it) before even looking at the logs.  Doing so would mean I would forego the retailer's exchange period in hopes of DJI fixing/replacing the drone.  Based on others' negative experience with DJI customer service and an extreme reluctance to repair/replace defective drones at their own cost, I thought it best to exchange it at the retailer.  Hopefully, this one doesn't give me any trouble, but if it does I will keep bringing them back until I get one that doesn't fall out of the sky for seemingly no reason (or until the retailer gets tired of replacing them).

KSBV Posted at 2018-5-1 15:24
Open a case on this one.  Your battery stopped reporting remaining capacity after it hit 82% and my guess is that it died in flight and the app didn't know it was that low and RTH wasn't prompted or initiated.  This was probably noticeable on your screen along with the longer than normal flight time but guessing you missed it while trying to fly/record/etc.

At least you proved that the Spark really does have a 15 minute flight time!

I opened a case, and from DJI technical support told me, they would replace my drones with a new one. I love DJI support

My DJI Spark felt out of sky into water. RTH feature was activated and it slowly declined then just felt into the lake. Send it to DJI. DJI said it hit something while on RTH and that caused the fall. Its clear (even on the map) that it was coming back, then declining in altitude over the lake and then just felt. Very dangerous, definitely it could hit anything there as seen on the map. DJI said its pilots fault and sent a bill for $457.00 (when new Spark retails for $349 at BHP) Btw they charged for battery too which was sent to DJI and was fine. No more DJI products, thanks!

My DJI Spark felt out of sky into water. RTH feature was activated and it slowly declined then just felt into the lake. Send it to DJI. DJI said it hit something while on RTH and that caused the fall. Its clear (even on the map) that it was coming back, then declining in altitude over the lake and then just felt. Very dangerous, definitely it could hit anything there as seen on the map. DJI said its pilots fault and sent a bill for $457.00 (when new Spark retails for $349 at BHP) Btw they charged for battery too which was sent to DJI and was fine. No more DJI products, thanks!

AlexNo Posted at 2018-6-15 11:23
My DJI Spark felt out of sky into water. RTH feature was activated and it slowly declined then just felt into the lake. Send it to DJI. DJI said it hit something while on RTH and that caused the fall. Its clear (even on the map) that it was coming back, then declining in altitude over the lake and then just felt. Very dangerous, definitely it could hit anything there as seen on the map. DJI said its pilots fault and sent a bill for $457.00 (when new Spark retails for $349 at BHP) Btw they charged for battery too which was sent to DJI and was fine. No more DJI products, thanks!
[view_image]

Hi,

If you want independent opinion, please upload your flight log here: http://www.phantomhelp.com/LogViewer/.
Forum members may take a quick look and see if we can figure out what happened.

Mirek6 Posted at 2018-6-15 12:25
Hi,

If you want independent opinion, please upload your flight log here: http://www.phantomhelp.com/LogViewer/.

Hi, thanks!
Just did http://www.phantomhelp.com/LogViewer/U6JNQZR9PHDDPPC0AE0S/

Seems like aircraft lost correct VPS altitude

AlexNo Posted at 2018-6-20 18:41
Hi, thanks!
Just did http://www.phantomhelp.com/LogViewer/U6JNQZR9PHDDPPC0AE0S/

This case is controversial.

Spark was closer to you than 30 metres.
Your RTH altitude was set to 30 metres (your "Return at current altitude" option was not enabled).
In such case Spark's programmed and correct behaviour is to land where it is. In this case - in the water.

Sparks (and not only Spark's) VPS sensors may be fooled by water reflections. In this case Spark thought that it was higher than it really was. It thought that it was landing safely while it was not.

Experienced pilot reaction would be to yank Spark up by manual action. I don't know what you did because stick movement data is missing. Perhaps you just froze in terror looking at Spark plunging in the lake.

So:
Manual clearly describes Spark's RTH behaviour - it was correct. DJI may expect that you should have known that Spark would be landing where it was.
Issues with Spark flying low over water are very well documented on various forums but I am not sure if the warning is in Spark's manual. Check it out. DJI does recommend not to fly lower than 2 metres above water due to VPS sensors confusion. But - if it is not clearly stated in the manual, you may have a case for DJI's negligence in informing you about this possible danger versus pilot's error.

AlexNo Posted at 2018-6-20 18:41
Hi, thanks!
Just did http://www.phantomhelp.com/LogViewer/U6JNQZR9PHDDPPC0AE0S/

I double checked the manual.
The chances are - you have no case.

One correction to my earlier post: Spark will start to auto-land when RTH is hit and "RTH at current altitude" is not set when Spark is up to 20 metres from home point (not up to 30 as I stated). However, looking at maps, your Spark was within that range of home point so auto-landing upon RTH was correct bahaviour.

I found warnings about flying over water in Spark's manual. While DJI does not state what minimum altitude is recommended (I do recall their forum moderators talking about minimum 2 metres), they do warn to be vigilant. See picture below.

In summary: I feel your pain. The warnings perhaps should be clearer but:
- You made a mistake by initiating RTH at closer distance than 20 metres from home point while Spark was over water.
- You did not know about DJI warning in the manual about risks of flying over water.
- You most likely did not react while Spark was landing (this I cannot confirm due to missing data but DJI may have additional data from logs which they perhaps did see your inaction).

AlexNo Posted at 2018-6-20 18:41
Hi, thanks!
Just did http://www.phantomhelp.com/LogViewer/U6JNQZR9PHDDPPC0AE0S/

Hey AlexNo,

One last question just for completeness. Did you use Remote Controller or your mobile phone to control Spark? As I mentioned - RC stick movement data is missing from the logs.

Mirek6 Posted at 2018-6-21 07:23
Hey AlexNo,

One last question just for completeness. Did you use Remote Controller or your mobile phone to control Spark? As I mentioned - RC stick movement data is missing from the logs.

I used phone as controller; it lost signal from Spark and then when spark was in fail RTH mode; then it got signal back via wifi and in 4 seconds felt with very high velocity
I get the point that it was in RTF radius, but I still expected to land softly not like a piece of rock. Even visual was distracted with water, I still expected it could rely on GPS positioning which is -+0.5m not

Thats what DJI said: "At t=6:23's , relative height=-22m, unit recorded external impact, and fell."

And then the fact that DJI estimated me with a bill of $450 is ridiculous when they sell the same package online for $399.

Not to rub salt in the wound, but I would not fly (take off) with only 10 Satellites...   or ever fly over water.  Sorry for your loss.

AlexNo Posted at 2018-6-21 10:34
I used phone as controller; it lost signal from Spark and then when spark was in fail RTH mode; then it got signal back via wifi and in 4 seconds felt with very high velocity
I get the point that it was in RTF radius, but I still expected to land softly not like a piece of rock. Even visual was distracted with water, I still expected it could rely on GPS positioning which is -+0.5m not

AlexNo,

Yes – I suspected that you used your phone not RC (based on lack of stick movement data) – thanks for confirming that.

I looked more closely based on what DJI said to you.
Your barometric pressure altitude sensor and VPS sensor got out of whack very early in the flight.
Based on the data I see there seems to be steep descent to the pond of, perhaps, ~20 metres vertical descent? - please confirm.

Baro and VPS sensors got out of whack 15 seconds from start of your flight when you were still over the ground – not water. VPS shows altitude of 2 metres lower than barometric pressure sensor.
At 30.3 seconds, they are already 4 meters out of whack (VPS shows 5.9 metres while barometric 11.9 metres). You are still over ground – not water.
15 seconds later the difference is 13 metres. VPS = 4.5 metres altitude while Baro = 17.4 metres altitude. Which one is correct? You are still over the ground so no messing up with VPS by water.
For the next one minute VPS is getting confused – shows gaps in data – altitude jumps. You are still above the ground.

My bet here is that VPS shows wrong altitude and you were flying quite high – over 20 metres – please confirm.

At 2 min 20 sec you start lowering altitude. Still above the ground. VPS shows 5 metres while barometric shows 20 metres.

At 2 min 54 seconds Spark is at the level you are standing (altitude 0), however it is over the pond. Assuming that there is a decline and barometric pressure sensor is showing correct altitude until the end, it is probably 20 to 28 metres above the pond.

I will not analyze any more until I here your confirmation of my two statements above.
The fact is that baro and VPS altitude sensors are out of whack from the early start of flight.
The fact is that baro shows at some point that Spark is over the pond but 28 metres below you, yet when it hits something (water?) it is only 22 metres below you and over the same pond.
Either it hit the mast of the sailing boat and baro altitude is correct, or it hit the water and sensors are totally confused.

At 5m 38 seconds you lost communication with Spark while it was over you – not water. The communication loss was for about 40 seconds.
Puzzling – because it is so close to you. Should have had good signal.
20 seconds after losing signal with mobile Spark would start RTH.

When communication got re-established at 6m 18 seconds, Spark was already auto-landing and drifting over the pond. VPS altitude goes up while it is landing (reflections in the water confusing sensor?) while baro goes down.
It hits something (water? mast?) 1 second before communication ends. I see violent pitch and roll. VPS shows 5.5 metres of altitude above ground (water) level while baro shows 20 metres below where you stand.

As I mentioned, I do not see movements of your sticks so some of the above are speculation based on conflicting altitude data.
I need your confirmation first that my assumptions about Spark’s altitude and pond altitude are correct.

Mirek6 Posted at 2018-6-21 13:09
AlexNo,

Yes – I suspected that you used your phone not RC (based on lack of stick movement data) – thanks for confirming that.

Thanks for analysis!!

There no steep descent to the pond. It actually can be seen here https://www.google.com/maps/@42.1574572,-87.9398228,3a,75y,87.81h,93.84t/data=!3m6!1e1!3m4!1svzMxMU3BFehKrEu5ueRyUQ!2e0!7i13312!8i6656
No boats, masts etc in the pond. Descent to Pond is hardly 3m deep
Drone was flying very high, more than 20m in sky (I could hardly see it with my eyes and then when i lost control over it, I couldnt see it at all visually)
When I lost signal to spark it was above me, thats correct.
In last second it hits edge of the pond (thats why I could grab it right away)

Here is photo of the pond, I figured out google link to maps is not working

>> "Puzzling – because it is so close to you. Should have had good signal"

Alex - Note the nearby power lines (see page 46 of Spark User Manual).  There are better places to fly near Buffalo Grove.

AlexNo Posted at 2018-6-21 16:52
Here is photo of the pond, I figured out google link to maps is not working

[view_image]

AlexNo,

If I were you, I would escalate the claim with DJI again. Perhaps one of this forums DJI moderators can help you?

Perhaps there were some newbie pilot mistakes you made, as I pointed out in my post #547.

However, your Spark had problems from 15 seconds after start. Its altitude data was getting confused and it was getting out of whack as you flew over the ground. This should not have happened and does point to Spark’s malfunction (or perhaps your VPS sensor was covered with dust?)
Normal behaviour is that to about 8 meters altitude, when flying up, both barometric and VPS sensors are in sync. Higher than about 8 metres VPS normally shuts off and barometric takes over.
Its not what was happening with your Spark. At altitude of 20 metres above the ground, VPS shows 5 metres altitude.

When you lower Spark and fly over the pond, this out of whack arrangement holds. At some point barometric sensor shows 28 metres below where you started. This, looking at the picture, is false reading.
At this time I believe you were flying on VPS sensor but with VPS sensor messed up from the very start.
DJI may say that VPS was fooled by the water and thought it was higher while it was very close to the water. But VPS sensor showed erroneous measurements from the very start when you were flying above the ground.

DJI’s conclusion is flat wrong. Yes – you hit something (water) but not at -22 metres but perhaps (again – looking at the picture since altitude data you have in logs in unreliable) at -3 to -5 metres.
They did not do their analysis correctly, but, I have seen it time and again, that with this forum moderators help, you can get the case reopened and re-analyzed.

The loss of signal is unfortunate, but could be contributed to anything, perhaps you mobile, perhaps Spark. I do not see any obvious interferences around (lamp posts which I see on the picture should not interfere too much and you did not get interference warnings in logs).

Perhaps DJI can recover your stick movements from some other logs using methods only they know. I can’t see them.

Contact DJI again and be firm. Their support, in my experience, is quite good but sometimes you need to give it an extra push. Good luck!

marlowe Posted at 2018-6-21 19:30
>> "Puzzling – because it is so close to you. Should have had good signal"

Alex - Note the nearby power lines (see page 46 of Spark User Manual).  There are better places to fly near Buffalo Grove.

These are lamp posts.
Logs do not show any interference - this, in my view, did not cause any issues analyzed in the posts above.

Mirek6 Posted at 2018-6-22 05:32
AlexNo,

If I were you, I would escalate the claim with DJI again. Perhaps one of this forums DJI moderators can help you?

Thanks Mirek6! Very helpful! I contacted DJI to reevaluate they analysis

I felt something was wrong with altitude readings but couldn't really figure out the whole thing, now it makes sense. Normal behavior was when it lands it slows down around 0.5m above the ground and then lands. And in my unfortunate case it seems like it was full speed descending and then hitting water (also it hit water in very shallow place where a lot of dark stones, evening no reflections)

Thanks!

marlowe Posted at 2018-6-21 19:30
>> "Puzzling – because it is so close to you. Should have had good signal"

Alex - Note the nearby power lines (see page 46 of Spark User Manual).  There are better places to fly near Buffalo Grove.

these are lamp posts near k1 speed facility in Buffalo Grove, shouldn't interfere with the drone. But I agree there better places to fly at Buffalo Grove

AlexNo Posted at 2018-6-22 06:36
Thanks Mirek6! Very helpful! I contacted DJI to reevaluate they analysis

I felt something was wrong with altitude readings but couldn't really figure out the whole thing, now it makes sense. Normal behavior was when it lands it slows down around 0.5m above the ground and then lands. And in my unfortunate case it seems like it was full speed descending and then hitting water (also it hit water in very shallow place where a lot of dark stones, evening no reflections)

The main thing is to show disagreements between two altimeters from the very start - second 15 onward. This may point at some internal malfunction which ultimately confused Spark into crashing into water. It was only getting worse as your flight progressed with total confusion of altimetres.

Do not argue about the confusion low above the water. This would be normal and this is what DJI warns about. In any case, trouble started early and you received no warnings.

Keep us posted on your progress with DJI claim. Ask them to get DJI moderators involved and ask them to look at what forum analysis showed. You may have a case.

Mirek6 Posted at 2018-6-22 05:34
These are lamp posts.
Logs do not show any interference - this, in my view, did not cause any issues analyzed in the posts above.

There are huge hi-tension power line towers just east of the pond (at least as shown on Google maps-see next post below). I have been to this area. It is not advisable to fly there.