Hi
I was flying my DJI Spark and I was experiencing signal errors, I tried returning to home and on it's way back, it entered ATTI MODE. I then lost signal from the RC and have not been able to locate it. I tried going to the last known location but I couldn't see it, it was quite high up. Flight log in the link below. I notice it says YAW ERROR in the log, it didn't say anything about YAW ERROR in the app while I was flying. There were also sufficient sattelites.

http://www.phantomhelp.com/logviewer/R3OLOWW6MEIAMMGX8B5P/

Interesting...waiting to read what others have to say.  Though it does seem like you were flying in a highly wifi congested area

Tviscomi Posted at 2018-1-20 05:35
Interesting...waiting to read what others have to say.  I've seen a lot of these errors over on the Spark forum, but not too many here.

Hi, thanks! Hopefully I can find out what went wrong. When you say you’ve seen it on the spark forum, where is that? I thought this was the spark forum.

KabirTheSpark Posted at 2018-1-20 05:37
Hi, thanks! Hopefully I can find out what went wrong. When you say you’ve seen it on the spark forum, where is that? I thought this was the spark forum.

I edited my original post

Whatever it maybe, file with DJI support concurrently.

rimza Posted at 2018-1-20 05:53
Whatever it maybe, file with DJI support concurrently.

Thanks! Where can I do that?

bottom of the web page, click support and go from there.

BTW where do you purchase it?

rimza Posted at 2018-1-20 06:02
bottom of the web page, click support and go from there.

BTW where do you purchase it?

I bought mine from ORMS. It’s our photographic retailer in South Africa

Sorry to read about your loss - It will be interesting to see what caused this.

I won't be able to explain everything, but here are my observations. There was a very similar case recently, even in the same country. I still don't know if the regulations in this country allow flying over residential areas like this one, but two issues come along with doing it. First of all many potential sources for interference e.g. through other WiFi networks and impossible to fly in VLOS at all times. Thus connection losses are almost certain to happen as it also was the case here.
The first one occured at 1m 38.8s which triggered Auto RTH for Spark as it is expected to happen. I do not know what triggers the "Command Timeout" error but at 2m 5.5s RTH stopped (or was manually interrupted by pilot) and Spark proceeded flight with stable GPS link and RC connection.Starting from 2m 15s maximum permitted altitude was exeeded considerably and soon Sport Mode engaged. Spark was then flown to the maximum controlled distance of this flight which was 385 meters in a built up area. Yaw errors started at 3m 13.7s. At 3m 14.9s the first compass error occurred and Spark switched into ATTI mode. From this point onwards RTH was no longer an option, only manual flight could have brought it home. But due to the distance, extremely high altitude and lack of VLOS this was not really a very promising option either.
The flightrecord clearly shows that the Spark was carried away by the wind and distance further increased up to 402 meters. At 3m 37.3s GPS link was re-established, after another connection loss at 3m 59.5s Spark went again into Auto RTH and after new yaw errors back into ATTI mode at 4m 10.5s.
The flightrecord then ends at 4m 15.5s, probably due to RC connection loss. Since at that time it was at almost 156 meters altitude it is unlikely that it has crashed into something at that point.
I am also sorry about your loss but you should accept that unsuitable flight environment, lack of VLOS and excess of altitude have been major contributing factors for it.
I recommend opening a support case with DJI here https://www.dji.com/support but am not able to predict the outcome.

Wachtberger Posted at 2018-1-20 08:38
I won't be able to explain everything, but here are my observations. There was a very similar case recently, even in the same country. I still don't know if the regulations in this country allow flying over residential areas like this one, but two issues come along with doing it. First of all many potential sources for interference e.g. through other WiFi networks and impossible to fly in VLOS at all times. Thus connection losses are almost certain to happen as it also was the case here.
The first one occured at 1m 38.8s which triggered Auto RTH for Spark as it is expected to happen. I do not know what triggers the "Command Timeout" error but at 2m 5.5s RTH stopped (or was manually interrupted by pilot) and Spark proceeded flight with stable GPS link and RC connection.Starting from 2m 15s maximum permitted altitude was exeeded considerably and soon Sport Mode engaged. Spark was then flown to the maximum controlled distance of this flight which was 385 meters in a built up area. Yaw errors started at 3m 13.7s. At 3m 14.9s the first compass error occurred and Spark switched into ATTI mode. From this point onwards RTH was no longer an option, only manual flight could have brought it home. But due to the distance, extremely high altitude and lack of VLOS this was not really a very promising option either.
The flightrecord clearly shows that the Spark was carried away by the wind and distance further increased up to 402 meters. At 3m 37.3s GPS link was re-established, after another connection loss at 3m 59.5s Spark went again into Auto RTH and after new yaw errors back into ATTI mode at 4m 10.5s.

Thanks for your observations. I’ve logged it with DJI. I’ll hopefully get feedback next week sometime. I was flying higher to try and maintain VLOS (I find that easier when the drone is a bit away). I will be more cautious if I ever get it back! Also won’t be relying on its GPS system too much in future

KabirTheSpark Posted at 2018-1-20 09:49
Thanks for your observations. I’ve logged it with DJI. I’ll hopefully get feedback next week sometime. I was flying higher to try and maintain VLOS (I find that easier when the drone is a bit away). I will be more cautious if I ever get it back! Also won’t be relying on its GPS system too much in future

I wish you good luck with the outcome and happy landings in the future. Just one last comment, the GPS system was not the problem in this case.

My mind is really blown every time I look at one of these flyaway logs. Especially those that have completely unexplained errors. When I look closely at the log I cannot see anything wrong with the actual data. When the Yaw Errors occur and the drone enters ATTI mode the telemetry data is still 100% accurate. The yaw position doesn't change and the drone seems to be continuing the point the same direction as seen in similar reports where the pilot has uploaded cached video.

I just find it hard to believe that this is an issue with interference when the location to home and all other telemetry data seems to have zero discrepancy. Heck, even the speed is still being reported when in ATTI mode, which I presume is calculated via GPS. Considering the number of recent cases with similar errors where DJI has deemed it as a full warranty repair I am beginning to suspect that this is either a software issue or something inherit to the hardware design of the Spark that DJI simply does not wish to address. (I'm more convinced it is a hardware defect as many report not having any issues, so it might be isolated to a specific vendor that is supplying the parts for some of the Sparks)

Oh well, this is purely speculation on my part, but @KabirTheSpark please let us know what DJI says.

asourceunknown Posted at 2018-1-20 18:48
My mind is really blown every time I look at one of these flyaway logs. Especially those that have completely unexplained errors. When I look closely at the log I cannot see anything wrong with the actual data. When the Yaw Errors occur and the drone enters ATTI mode the telemetry data is still 100% accurate. The yaw position doesn't change and the drone seems to be continuing the point the same direction as seen in similar reports where the pilot has uploaded cached video.

I just find it hard to believe that this is an issue with interference when the location to home and all other telemetry data seems to have zero discrepancy. Heck, even the speed is still being reported when in ATTI mode, which I presume is calculated via GPS. Considering the number of recent cases with similar errors where DJI has deemed it as a full warranty repair I am beginning to suspect that this is either a software issue or something inherit to the hardware design of the Spark that DJI simply does not wish to address. (I'm more convinced it is a hardware defect as many report not having any issues, so it might be isolated to a specific vendor that is supplying the parts for some of the Sparks)

Thanks! I hope to hear from them during the week. Will keep this thread updated. A similar thing happened to a friend of mine with a Mavic. It flew away

I am terribly sorry for your loss, Kabir. You had contacted our support and reported this case, right? If so, they will help to submit a data analysis if the drone is under warranty period. If there is a need, you can leave me with your ticket or case number, I will help to check the case.

This happened to me while I was flying at about 700m it might be because of strong gusts

KabirTheSpark Posted at 2018-1-20 09:49
Thanks for your observations. I’ve logged it with DJI. I’ll hopefully get feedback next week sometime. I was flying higher to try and maintain VLOS (I find that easier when the drone is a bit away). I will be more cautious if I ever get it back! Also won’t be relying on its GPS system too much in future

the minute you change the default altitude limit to  a higher limit devoids the warranty

DJI Thor Posted at 2018-1-20 21:33
I am terribly sorry for your loss, Kabir. You had contacted our support and reported this case, right? If so, they will help to submit a data analysis if the drone is under warranty period. If there is a need, you can leave me with your ticket or case number, I will help to check the case.

Thank you! I have logged it with DJI support. This is my reference number - 947279

asourceunknown Posted at 2018-1-20 18:48
My mind is really blown every time I look at one of these flyaway logs. Especially those that have completely unexplained errors. When I look closely at the log I cannot see anything wrong with the actual data. When the Yaw Errors occur and the drone enters ATTI mode the telemetry data is still 100% accurate. The yaw position doesn't change and the drone seems to be continuing the point the same direction as seen in similar reports where the pilot has uploaded cached video.

I just find it hard to believe that this is an issue with interference when the location to home and all other telemetry data seems to have zero discrepancy. Heck, even the speed is still being reported when in ATTI mode, which I presume is calculated via GPS. Considering the number of recent cases with similar errors where DJI has deemed it as a full warranty repair I am beginning to suspect that this is either a software issue or something inherit to the hardware design of the Spark that DJI simply does not wish to address. (I'm more convinced it is a hardware defect as many report not having any issues, so it might be isolated to a specific vendor that is supplying the parts for some of the Sparks)

Yes, the speed is calculated using GPS data. The Spark LatLong position recorded is used, so what you see in the log is the actual groundspeed. (so not the fwd speed of the drone but the direction moving speed)
That why you read a positive speed value when the drone is flying backwards, i.e. when heading into strong wind in the GoHome mode.

Sorry wrong discussion.............

KabirTheSpark Posted at 2018-1-21 00:20
Thank you! I have logged it with DJI support. This is my reference number - 947279

Hi, I'd checked your ticket, our support is following up your case now. Please wait patiently.

If the problem is gps conflicting with compass because of certain manoeuvres, you never will see anything in the logs.
The telemetry will appear plausible, but may be inaccurate. DJI will know all this by now, and if they fix it,we'll never know.

DJI have offered a replacement drone for my flyaway.

Thanks for all your help!

Congratulations on good outcome! I bet you wont be relaying on the RTH next time =)

Good result. Would you be good enough to update us as to whether your new spark performs as it should? Hardly anyone ever does this.

Bright Spark Posted at 2018-1-24 02:35
Good result. Would you be good enough to update us as to whether your new spark performs as it should? Hardly anyone ever does this.

Will do! Well, if I remember. I'll try to I'm waiting for feedback from DJI on the problem. I want to know if something was wrong with my Spark or it's just something that can happen

KabirTheSpark Posted at 2018-1-23 22:15
DJI have offered a replacement drone for my flyaway.

Thanks for all your help!

Great to hear, congratulations!