KlooGee Posted at 2018-6-20 07:43
Hodgy, I'm sorry that you have lost your Spark in an unfortunate event.  Unfortunately, there isn't a lot to go on in your log.  I feel for you that you lost your aircraft, but I'm going to be honest with you here.

The main technology that DJI drones rely upon for safe navigation is GPS.  When you choose take off with no GPS locked in at all, then you exponentially increase the risk of crashing or losing your aircraft.  I know you were only hovering in front of you, but without its main navigation system being operational, if everything else doesn't go 100% correct, then you will be in a very bad situation.  

Interesting is why in the video app player GPS is shown....with zero satellites.
So what did the GoApp display showed during this short flight?

Mayby DJI can explain ?
cheers
JJB

Another DJI flight log showing P-GPS flight mode with 0 satellites and clearly in ATTI. Sigh ...

I see fairly chaotic stick movements - not much of extensive flying.
Spark was close.
I do not see consistency in attempted landing. Fairly chaotic quick bursts of Throttle stick, up, down, up, down.
Last two seconds look like crash - large and unusual pitch and roll swings which follow couple 4 seconds of full speed flying to the side. Almost as if Hodgy yanked the roll stick for 4 seconds sideways (46 sec to about 50 sec point) and the AC crashed into something.
Communication with AC lost 4 times at such short distance (unless 4 "downlink restored" messages are false positives). Weird.

So - logs kind of agree with what DJI says (in my opinion). Behavior Hodgy described is very, very unusual and not consistent with logs (unless it occurred after communication was lost).

JJBspark Posted at 2018-6-20 08:18
Interesting is why in the video app player GPS is shown....with zero satellites.
So what did the GoApp display showed during this short flight?

I noticed the same.  However, I didn't dive into the deep parts of the logs.  Is it possible it was in Opti mode when the log actually shows GPS mode?

KlooGee Posted at 2018-6-20 13:05
I noticed the same.  However, I didn't dive into the deep parts of the logs.  Is it possible it was in Opti mode when the log actually shows GPS mode?

yes, i think that is the case.

In the log there is a value if the VPS is in use (2values; 1) VPS height and 2) VPS "forward"), that correlates with OPTI mode i found out.
How, did some test flying into the forest where i lost GPS mode into OPTI into ATTI.
Analyzing that log made it easy to know now when OPTI is active, in my program its an option to logically show OPTI instead of GPS.

cheers
JJB

sorry about your loss man, but starting motors and lift off with no GPS is not good, even if it was to just hover a bit.. not sure what more can be done to assist you in this instance!

KlooGee Posted at 2018-6-20 07:43
Hodgy, I'm sorry that you have lost your Spark in an unfortunate event.  Unfortunately, there isn't a lot to go on in your log.  I feel for you that you lost your aircraft, but I'm going to be honest with you here.

The main technology that DJI drones rely upon for safe navigation is GPS.  When you choose take off with no GPS locked in at all, then you exponentially increase the risk of crashing or losing your aircraft.  I know you were only hovering in front of you, but without its main navigation system being operational, if everything else doesn't go 100% correct, then you will be in a very bad situation.  

Hi KlooGee,

Thank you for taking the time to look at this. And yes I can see it from DJI's perspective, and with this type of product there's going to be a lot of complaints that they have to be strict with.

Hopefully I can get back in the air, one way or another.

Thank you

JJBspark Posted at 2018-6-20 22:15
yes, i think that is the case.

In the log there is a value if the VPS is in use (2values; 1) VPS height and 2) VPS "forward"), that correlates with OPTI mode i found out.

Hey thanks for looking at this.

I am not very good with the technical side, is this something worth looking into? or just an irrelevant detail
?

kind regards

Mirek6 Posted at 2018-6-20 10:59
Another DJI flight log showing P-GPS flight mode with 0 satellites and clearly in ATTI. Sigh ...

I see fairly chaotic stick movements - not much of extensive flying.

Hi Mirek6

Thank you for the detailed analysis and looking into this. I am not an expert with this kind of thing and the flight was over 2 months ago but in regards to your comments:

I would say the 'fairly chaotic throttle bursts' throughout the flight were me just playing around as the drone was meters in front of myself, enjoying/learning it's response.

As for the final stick movements (in mode 2) The drone began to act extremely erratically, (not just horizontal drifting - which is expected in ATTI mode), and then shot off into the distance, - it must have definitely disconnected from the log as apparently there is no sign of this, or my drone ) :.

I really don't know why it shows (so I'm told) that it crashed with obstacles / flipped in front of myself, I would just pick it up then and fix any damage - which would surely be minimal at that speed.

When you say 'communication was lost 4 times' is this normal behavior, or something worth mentioning?

Is there also a way to view stick movements after the end of the flight log, as it would show me frantically trying to pull the drone back? - It seems strange to end so abruptly?

Anyway, Thank you for your time

Hodgy

JJBspark Posted at 2018-6-20 22:15
yes, i think that is the case.

In the log there is a value if the VPS is in use (2values; 1) VPS height and 2) VPS "forward"), that correlates with OPTI mode i found out.

Thank you JJB for shedding more light on what shows in logs as AC mode and what mode AC really is (OPTI vs P-GPS). This is very important finding which confirms my conviction (developed in the past few days) not to take this data at face value but rather be more vigilant and check against everything else. We have seen that DJI logs may show P-GPS while AC was most likely in ATTI. Now we know that it can show P-GPS while the AC was most likely in OPTI.

Mirek6 Posted at 2018-6-21 05:07
Thank you JJB for shedding more light on what shows in logs as AC mode and what mode AC really is (OPTI vs P-GPS). This is very important finding which confirms my conviction (developed in the past few days) not to take this data at face value but rather be more vigilant and check against everything else. We have seen that DJI logs may show P-GPS while AC was most likely in ATTI. Now we know that it can show P-GPS while the AC was most likely in OPTI.

well, i think i never saw a file wich shows GPS while in thought i should be ATTI while GPS count and reception were there.
But that`s just a feeling.
In this last file GPS count was ZERO and in GPS mode, wich is really strange....

Hodgy_spark Posted at 2018-6-21 00:55
Hi Mirek6

Thank you for the detailed analysis and looking into this. I am not an expert with this kind of thing and the flight was over 2 months ago but in regards to your comments:

Hodgy,

Last few seconds before crash:
Sec 44.9: Altitude about 1.7 metres. For 1.5 seconds you pull your throttle down directing Spark lower. It responds correctly.
Sec 46.4: Altitude about 1.4 metres. You stop directing Spark down. Your throttle (altitude stick) gets back to neutral.
However, half a second earlier you directed Spark full speed to the side. Spark responds, gets bit wobbly for few tenths of a second (gust of wind?)
Sec 49.4: Spark's roll starts to be slightly inconsistent with your aileron stick (you are directing Spark to one side when it starts tilting to the other). Consistent with small air disturbances close to trees? Some obstacle? You also put throttle down for half a second.
Sec. 49.8: Start of some event (crash?): Spark's altitude according to VPS is 70 cm. You yank on you throttle up and speed forward. This lasts less than a second while Spark is heaving extensively to one side and the other in a span of second. This may be crash or Spark got into shrubs or some kind of foliage which tosses it violently around.
VPS altitude is erratic as if Spark went down, hit the ground and flew a bit up. Perhaps it flipped in the air which confused VPS. In any case – something unexpected and violent happened.
Sec. 50.9: Log ends with no sign of Spark’s fly away.

To answer your “downlink interrupted” question. We see it often in logs. Your communication with Spark was interrupted 4 times: for 1 second. 1.4 seconds, 1.7 seconds and 2.4 seconds.
With Spark flying close to you it is, in my view, unusual.   

JJBspark Posted at 2018-6-21 05:10
well, i think i never saw a file wich shows GPS while in thought i should be ATTI while GPS count and reception were there.
But that`s just a feeling.
In this last file GPS count was ZERO and in GPS mode, wich is really strange....

JJB,

You say: "well, i think i never saw a file which shows GPS while in thought i should be ATTI while GPS count and reception were there."
How about recent case in Turkey of Spark flying away under high power lines?  https://forum.dji.com/thread-152847-1-1.html
You were involved in analysis of this case.

At first I concluded "true fly-away" based on strong GPS showing and P-GPS mode. KlooGee on the other hand, concluded ATTI fly-away pushed by prevailing EEN winds. I do lean towards his conclusion - see exchange of post between KlooGee and me:
Post #34 with KlooGee analysis (different than mine)
My post #42 and than #47 asking question about P-GPS showing while AC was in ATTI.
KlooGee replied with post #53 confirming my observation with his experience as well.

What is your opinion?

Hi, good question....

I think that the ATTI en GPS mode were correct. In ATTI drone drifted away ~ 275 ~ heading, no RC input. red move line and drone purple heading line alignd.
GPS start to work, RC yaw input had effect on drone, turned toward home but too far, distance decrease in increase again.
At 100 meter away Spark is hard to see, watching againt a green hill. Happend all in 30 secs.

So IMO not a fly away on its own, but ATTI mode followed by GPS mode, too difficult to make quick the right steering inputs. Close to the hill so it crashed.

BTW to be clear about my comment ; i am not in any way say or think that i could fixed this situation. (mayby yes. mayby not)
cheers
JJB

JJBspark Posted at 2018-6-21 06:47
Hi, good question....

I think that the ATTI en GPS mode were correct. In ATTI drone drifted away ~ 275 ~ heading, no RC input. red move line and drone purple heading line alignd.

JJB,

I like discussions with you - they are always constructive.
I hope that forum members forgive us discussing different case on this thread but I think that concluding it is important to our understanding of DJI logs inconsistency.

I can't interpret your pictures since I do not know your program (I will search for it on this forum soon and if I can't find it, I will ask you for a link :-))

You are right in your observations of Spark's behaviour until it started to fly away. Spark, despite many errors and dropping into and out of ATTI behaved consistently with its pilot RC stick movements. But the true problem begins at 3m 42.6 sec. Here, logs show P-GPS. Suddenly Spark takes off (from almost standstill) in the direction of WWS at fairly high speed with no input from controller. 13 seconds later it crashes. During entire flight there was no input from sticks.

This has only one of the two possible explanations:
1. At 3m 42.6 sec with GPS on Spark decided to return to wrong home. Pilot did not stop it nor did he do anything. I called it "true fly-away". It is possible if h/w or s/w malfunctions.
2. At 3m 42.6 sec Spark dropped to ATTI mode and, without any input from sticks, got pushed by fairly strong EEN wind in WWS direction. This was KlooGee's findings based on the fact that earlier in the logs, when the Spark was holding GPS position in one spot, we could see its pitch and roll angles consistent with fairly strong wind blowing from EEN. Spark's behaviour is fine - no malfunction.

Flying for 13 seconds in P-GPS before crash with high speed with no stick input? If Spark was OK and in P-GPS, it would hold position - as it was earlier in the flight - and not fly on its own.

In summary, I believe KlooGee was right (scenario 2 above). Logs show wrongly P-GPS while Spark was in ATTI. Wind's speed and direction, ATTI mode and no input from pilot fully explain last 13 seconds of Spark's flight before crash. Logs show wrong mode.

See my picture below:

Mirek6 Posted at 2018-6-21 05:50
Hodgy,

Last few seconds before crash:

Hi Mirek6,

Thank you for your detailed analysis, which reflects DJI's outcome  and clearly the available flight data  too.

It definitely didn't crash into anything (at least as far as I could see), it did start acting very erratically as you describe, and did at least tilt a lot, not sure about flipping though, when I would imagine the fly away occurred....And for the stick movements it was like controlling a drunk person,(more towards the end of the available data), but I was not worried about crashing, as it was at least head height and not near any obstacles/floor

I can obviously see that my experience/points are not actual evidence, and the lack of area mapping does not help me in anyway, as I can't show the safe, flat, shrubless area etc... And no sign of fly-away is not good too + all the other factors not helping my case.... I also really like the flight record method and DJI's/community's approach to such cases.

Anyway, I have been persistent as what I personally witnessed I believe should not have occurred with such a product, as a simple hovering resulted in a erratic flyway etc. etc.

But, yes, thank you very much for your time and providing new, interesting information on my case. DJI have stopped answering any of my questions..

Kind regards

Hodgy

JJBspark Posted at 2018-6-21 05:10
well, i think i never saw a file wich shows GPS while in thought i should be ATTI while GPS count and reception were there.
But that`s just a feeling.
In this last file GPS count was ZERO and in GPS mode, wich is really strange....

Hi JJBspark,

Thank you for your analysis and interesting points. Yes all the 'GPS signal weak' , "gps  weak fly with caution' and GPS mode icons etc. showing, while there was  0 satellites, is/was misleading, as I assumed it would quickly get a stronger signal while hovering close. - It should have a much bigger, clearer warning sign, and maintain ATTI mode throughout...

Anyway, your other points/discussions of other cases are welcomed and interesting - as there is only so much my short flight record shows ) :

Thank you

Kind regards

Hodgy

Mirek6 Posted at 2018-6-21 16:34
JJB,

I like discussions with you - they are always constructive.

Hi mirek,

Think you are right, without RC input after 3.42 it increased speed thus fly away.
At that tine there was a CompassErrorLarge present (not shown in the GoApp) so beacuse of this error Spark did not hold position.
Thanks for activating my brains in a better way!  

cheers
JJB

Hodgy_spark Posted at 2018-6-21 19:19
Hi JJBspark,

Thank you for your analysis and interesting points. Yes all the 'GPS signal weak' , "gps  weak fly with caution' and GPS mode icons etc. showing, while there was  0 satellites, is/was misleading, as I assumed it would quickly get a stronger signal while hovering close. - It should have a much bigger, clearer warning sign, and maintain ATTI mode throughout...

HI Hodgy,

No thanks at all.

Think DJI should give pre-warnings in the App if there are compass and/or yaw and/or *.* errrors popping up. No you get a msg when its too late (exiting GPS mode....), i want earlier a msg and an audio tone!!

How often do we miss messages on the screen when watching the bird fly?? So DJI gives us more AUDIO!

If DJI moderators are reading this : please give a reaction.

cheers
JJB

JJBspark Posted at 2018-6-21 22:56
Hi mirek,

Think you are right, without RC input after 3.42 it increased speed thus fly away.

JJB,

You say: "Thanks for activating my brains in a better way! "

Yes - this is what I tend to do :-) And I love when others activate my brain in a better way - which happens quite often :-)

JJBspark Posted at 2018-6-21 23:03
HI Hodgy,

No thanks at all.

Yes, it would be great if there would be an option to add loud audio for error messages.

Mirek6 Posted at 2018-6-20 10:59
Another DJI flight log showing P-GPS flight mode with 0 satellites and clearly in ATTI. Sigh ...

I see fairly chaotic stick movements - not much of extensive flying.

Hi Mirek6, (and others)

Sorry to drag up an old post, it just takes DJI so long to respond to me....
I was wondering in regards to your point about the "communication lost 4 times at such a short distance".

i wanted to ask DJI about this, (maybe you figured it out to be normal).

Anyway, I mentioned it in my latest email to them with a few other questions.  and got the response 1. "Kindly note that no remote controller lost is verified in the flight record."  - (DJI)

I was wondering if there is any evidence I can present to them, to get a more straight forward answer?

( I also asked about why P-GPS etc. is shown when there were clearly no satellites and it should be in ATTI mode, and I got the response:

2. The aircraft worked under Opti mode after took off due to the weak GPS signal. The aircraft entered Atti mode as environmental conditions changed. The aircraft and app would need time to respond to the fight environment of your choice. The pilot could see the GPS signal in APP. Please refer to the data analysis report attached for reference. Thanks for your understanding. )

Anyway, if you (or anyone) has any good suggestions of questions / evidence to ask DJI regarding my case it would be much appreciated...

Thank you

Hodgy - spark

Hodgy_spark Posted at 2018-7-3 03:49
Hi Mirek6, (and others)

Sorry to drag up an old post, it just takes DJI so long to respond to me....

Hodgy,

You say: “Anyway, I mentioned it in my latest email to them with a few other questions.  and got the response 1. "Kindly note that no remote controller lost is verified in the flight record."  - (DJI)”

This statement is probably true. I have no way of verifying if up-link between Remote Controller and Spark was lost. As a matter of fact, from my analysis it looks like Spark was responding to your commands throughout its short flight.

However, It appears (through the missing logs and DJI Go messages) that down-link from Spark to Remote Controller was lost many times. Your log has more holes in it than swiss cheese :-). DJI GO reported successful restoration of down-link at: 12.8 sec., 14.5 sec., 24.8 sec., 26.5 sec., 30.1 sec. There are more interruptions in logs (which I see as missing rows – logs are entered every 0.1 sec. and you are missing plenty). Something was definitely not quite right and very unusual with Spark reporting its status to Remote Controller.

But… just speculating – if your mobile was busy with some other task(s) and could not process some of the information Spark was sending it, DJI GO could have interpreted it as a down-link failure. I do not know if this is possible but, theoretically, it is. Only DJI can say.

You say (DJI response): “The aircraft worked under Opti mode after took off due to the weak GPS signal. The aircraft entered Atti mode as environmental conditions changed. “

This is true (although Spark entered ATTI only when it crashed – this would be, in my opinion, normal due to mismatch of many sensors).

In any case, outside of possible downlink interruptions, which are unusual, logs confirm what DJI says. Logs are also consistent with a crash at 49.8 sec. Logs do not show any fly-away. They simply end with a crash.

If you still insist that this is not what happened (despite DJI findings and what I see in the logs) I would be interested in seeing entire analysis report which you got from DJI.
I would also be interested in seeing a video from this flight (it is cashed on your mobile). Otherwise you cannot win this case.

Mirek

Mirek6 Posted at 2018-7-3 07:18
Hodgy,

You say: “Anyway, I mentioned it in my latest email to them with a few other questions.  and got the response 1. "Kindly note that no remote controller lost is verified in the flight record."  - (DJI)”

Hi Mirek,

Thank you for your detailed response. The DJI's brief data analysis is below in screenshots.

I'm not sure how to view the cashed video from the flight (which I would be very interested in watching too) - On the flight log, the flight in question does not have an image/video next to it (screenshot attached), which would suggest there isn't one?

Overall, yes I do feel hopeless about this case, as I just get the same automated response from DJI about flying in ATTI mode (which as you point out, it only reverts to the moment at the end of the flight data), but yes I watched the spark flyaway so I have persisted...








And the screen recorded flight log is at


Many thanks

Hodgy_spark Posted at 2018-7-4 02:50
Hi Mirek,

Thank you for your detailed response. The DJI's brief data analysis is below in screenshots.

Hodgy,

You can watch your cached video as follows:
1. Start DJI GO.
2. Click Editor (bottom row).
3. Click Album (top row).
You will be able to find your cashed video there.

I do not know where to look for the video physically on iPhone but on Android device your videos are in the following folder:
Internal storage / DJI / dji.go.v4 / DJI_RECORD

You will see all your videos there. You can watch it or copy it to your PC.

Unfortunately for you, analysis which DJI provided does, in general, match my analysis of your logs. What I differ on is that your craft was in OPTI until almost the moment of crash and it did crash a few seconds before DJI indicated that it did. But these differences are minor and do not change overall situation.
Your only saving grace may be proof from the video that your drone did not crash and just flew away.

Good luck.

Mirek

Mirek6 Posted at 2018-7-4 18:50
Hodgy,

You can watch your cached video as follows:

Hi Mirek,

Thank you for all your help, it's really great to get clear answers from someone, rather than waiting days for blunt replies for DJI (although overall I do understand their standpoint).

I will check for the video when I get home and load it up.

I was hoping you could help me with one last favor. I have lost hope of proving the fly away as the data ends so abruptly - although not having the drone is a constant reminder (so I have been looking for other avenues),
Did the so called 'crash/abnormality' start before ATTI mode icon shows (which shows right at the end of the video 0:51)? And is there a way to clearly show this?

Kind regards

Hodgy

Hodgy_spark Posted at 2018-7-4 19:08
Hi Mirek,

Thank you for all your help, it's really great to get clear answers from someone, rather than waiting days for blunt replies for DJI (although overall I do understand their standpoint).

Hodgy,

You ask “Did the so called 'crash/abnormality' start before ATTI mode icon shows (which shows right at the end of the video 0:51)? And is there a way to clearly show this? “

Yes. Below is a fragment of a log with relevant data for the last two and a half seconds.
I highlighted in yellow abnormal pitch and roll values which, in my mind, show start of a crash. Perhaps Spark was tangled in tree branches or something like that. The fact is – it was wobbling violently.
Red highlight may show impact – see pitch and roll values.

Logs show that Spark switched to ATTI mode at 51.6 sec. But do not rely on this. Very often Spark flight logs show P-GPS while Spark is already in ATTI so this may be false reading.

I do not know what DJI meant by switching to ATTI due to “changing environmental conditions”.
In order to switch from OPTI to ATTI, optical sensors at the bottom of Spark would have to be really confused (rapidly changing patterns or rapidly fading light).

In my interpretation your Spark was in OPTI until, perhaps about 1.5 seconds before logs end. See yellow highlight regarding perceived altitude as measured by visual positioning system sensors.
As of 51.3 VPS shows Spark climbing from zero to 60 cm – clear sign of confusion, which, in my view, led to switch to ATTI.

My interpretation is as follows:
•        At 49.8s at altitude of about 70 cm above the ground Spark ran into some vegetation causing violent sways.
•        Half a second later VPS is getting confused by swirling patterns or perhaps not enough light.
•        At 51.2s Spark shows another wobble. VPS reading is abnormal and your handling of sticks is quite chaotic (OK – it all is happening in the span of 2 seconds so you are probably randomly yanking on sticks trying to save Spark). I suspect that, at that point, Spark is already in ATTI.
•        At 51.9s There is sudden change in pitch and roll angle and logs end – consistent with hard crash knocking Spark out.

Mirek

Mirek6 Posted at 2018-7-5 05:44
Hodgy,

You ask “Did the so called 'crash/abnormality' start before ATTI mode icon shows (which shows right at the end of the video 0:51)? And is there a way to clearly show this? “

Thank you very much Mirek, I really appreciate all your clear, useful info.

(I tried looking for the cashed video, but it predictably wasn't there, as with all the flight-records with missing images next to them - seen in the screenshot below)

Thanks again

Hodgy_spark Posted at 2018-7-5 07:09
Thank you very much Mirek, I really appreciate all your clear, useful info.

(I tried looking for the cashed video, but it predictably wasn't there, as with all the flight-records with missing images next to them - seen in the screenshot below)

OK - perhaps you were not recording while flying. If you record your flight, video would have been there.
The snapshot you provided is from the Flight Record page. You need to go to Editor / Album to see video.

Mirek

Mirek6 Posted at 2018-7-5 07:44
OK - perhaps you were not recording while flying. If you record your flight, video would have been there.
The snapshot you provided is from the Flight Record page. You need to go to Editor / Album to see video.

Ah Ok,  yes sadly I was not recording the flight in question

Thanks again!

My Spark Flew away after 11 seconds of flight in a completely unobstructed and remote place.  At 500 feet away the Drone dropped reception to the controller and just flew away at 35 MPH never to be seen again. It could have flew into a highway traffic, kids in a park, path of an airplane. DJI looked at the flight data and said because it was in ATTI mode it was totally my fault. I feel this is ridiculous. They refuse to answer these questions:
1) Even thought it was in ATTI why did it loose connection to my controller in a remote area with no RF interference or physical obstructions?
2) There are audible warnings about taking off, landing, low battery, etc. but if you are at such a risk of  losing your Drone when unknowing it is in ATTI mode, why is there not a similar warning?
3) The area I was at had STRONG GPS signal which I tested, why will they not answer my question why the failure of the GPS was not the true problem. Search the GPS issues with Spark and see how many have had iGPS issues!
4) When the Drone lost connectivity to the controller, why did it not stop and hover as they said it should have done? I could have retrieved it had this happened.

I believe DJI should replace my Drone for 3 reasons:
a) GPS failure
b) Drone did not stop and hover when losing connectivity (thus could result in very dangerous consequences). Two DJI agents said it should have stopped by design.
c)  Inadequate warnings when flying in a dangerous condition they call ATTI mode

I look forward to DJI answering these points

genewSpark Posted at 2018-8-1 08:02
My Spark Flew away after 11 seconds of flight in a completely unobstructed and remote place.  At 500 feet away the Drone dropped reception to the controller and just flew away at 35 MPH never to be seen again. It could have flew into a highway traffic, kids in a park, path of an airplane. DJI looked at the flight data and said because it was in ATTI mode it was totally my fault. I feel this is ridiculous. They refuse to answer these questions:
1) Even thought it was in ATTI why did it loose connection to my controller in a remote area with no RF interference or physical obstructions?
2) There are audible warnings about taking off, landing, low battery, etc. but if you are at such a risk of  losing your Drone when unknowing it is in ATTI mode, why is there not a similar warning?

1) You should start your own thread
2) You should upload your flight record to phantomhelp - that is pretty much the only way people on this forum can help you

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

genewSpark Posted at 2018-8-1 08:02
My Spark Flew away after 11 seconds of flight in a completely unobstructed and remote place.  At 500 feet away the Drone dropped reception to the controller and just flew away at 35 MPH never to be seen again. It could have flew into a highway traffic, kids in a park, path of an airplane. DJI looked at the flight data and said because it was in ATTI mode it was totally my fault. I feel this is ridiculous. They refuse to answer these questions:
1) Even thought it was in ATTI why did it loose connection to my controller in a remote area with no RF interference or physical obstructions?
2) There are audible warnings about taking off, landing, low battery, etc. but if you are at such a risk of  losing your Drone when unknowing it is in ATTI mode, why is there not a similar warning?

Hello,

yes it sounds similar to my case...

I wrote hundreds of long, time consuming emails explaining everything, and DJI respond with poorly written emails that don't relate to anything I mention.

I'm pretty sure they will just continually highlight that you were in ATTI mode, and no warranty issued in ATTI mode, and just keep responding in that same way regardless.

If you have any other points to the case you could try and bring them up. and number the questions like you have done... other people on the forum are very helpful and have good insight into flight logs and were very helpful for me, as I'm not good with that technical side of things...

good luck, and sorry about your rubbish experience with them.

S.J Posted at 2018-5-3 17:07
The SPARK is not meant to fly at low altitude as it is prone to interference. On outdoors, GPS signals in theory should be present at 3 mtr or 4mtr height but in reality you must fly the drone at minimum 10 to 15 meter height for a better GPS  lock .

This is interesting, but what's about the RTH position taken with the drone on the ground?

ssylca44 Posted at 2018-8-2 06:14
This is interesting, but what's about the RTH position taken with the drone on the ground?

The RTH position is taken when there is a GPS lock.  Sometimes this dosen't happen and therefore it is recommended to do a manual RTH through the GO4 app menu during flight.

YES* THIS HAPPENED TO ME IN THE SAME MANNER!

THEY PUT THE BLAME ALL ON ME..

THEY HAVE HAD MY SPARK FOR TO WEEKS AND EVEN MADE TO PAY FOR REPAIRS!

I ASK MANY QUESTIONS ONLY TO BE TOLD IT MY FAULT..
WHEN THE SPARK HOES INTO ATTI MODE WE CAN NOT CONTROL THE SPARK AND OR IT WILL CRASH OR GLY AWAY!

MANY HAVING THE SAME ISSUE I HAVE FOUND ME INCLUDED!

I HAVE SPENT COUNTLESS HOURS ON THE PHONE WITH DJI AND HAVE GOT NO WHERE WITH THEM!

GOOD LUCK!!

Hi I purchased my spark in March 2018 and have had many fun and enjoyable safe flights, the spark was well behaved with no issues, I recently updated the software as instructed via DJIGo 4 app for both controller and spark which went well, the first flight I had after the update ended in disaster during the flight I received the following then spark just flew away, I tried everything to get control but nothing would stop it, after hours of searching I found it in a farmers field amazing with no damage, I have lost all confidence and won’t fly again until I find out what went wrong any help is appreciated.