Hello all,
Unfortunately I lost my DJI Spark two hours ago I do not know what went wrong, but as far as I can see from the FlightLogs the Spark decided to do her own thing.
I've downloadded the log from my smartphone and uploaded it to http://www.phantomhelp.com/LogViewer, where I get a table view of all events.
For the last event it shows "Auto Landing" - does anybody know what that means exactly? Will it land at the corresponding position or are there any other processes during this time like trying to return to home?

I'll definitely search again for the cutie tomorrow morning as well as contacting the support....from my point of view I did not make anything wrong and checking the log it seems the drone went crazy.


Thanks for your support!

Can you please post the link to your logfile reading on phantomhelp.com? Then I'll be able to tell you a bit more at least about what might have happened.But yes, "Auto landing" means it has landed where it has been at that time". Reasons can be various, therefore post logfile here please.

Thanks for your quick reply.
Find below the links to the logfiles:
http://www.phantomhelp.com/LogViewer/0U0SWN8DV5UHVDT4G2HW/#

http://app.airdata.com/main?flig ... _id=GENERALOverview

Does it show battery status? Cause if it dips below 10%, no matter where or what its doing, it land there and then.

Yes, crystal clear case. You did not pay sufficient attention to the battery status although Spark was at range limit at that time. If you had watched the upper part of your screen, you would have seen the warnings. At the end Spark has initiated landing because of low battery at the spot where it has been at that time. Good luck in finding it (I hope your contact details are written on it in case someone finds it) and don't forget to use the "Find my drone function" in the App.

Thanks for your reply.
Okay... So I can be "sure" that the drone went down (tried to land) at the last reported position.
As you might see from the log, there is a big tree. Does anyone know whether the drone lands no matter if there are obstacles like a tree?

Thanks!

Wachtberger Posted at 2017-9-11 12:35
Yes, crystal clear case. You did not pay sufficient attention to the battery status although Spark was at range limit at that time. If you had watched the upper part of your screen, you would have seen the warnings. At the end Spark has initiated landing because of low battery at the spot where it has been at that time. Good luck in finding it (I hope your contact details are written on it in case someone finds it) and don't forget to use the "Find my drone function" in the App.

Thanks for your analysis.
Of course I checked the status indicator on my display and I've also tried to bring the drone back to me, but as the straight line respectively tracks might indicate, the drone did not respond to any of my inputs.
Also the RTH function did not work properly... it failed directly after I've called the function.
I wonder about the fact that the drone flew away this long distance without receiving any input. If I'm informed correctly the drone should return back automatically after the signal is lost...

I'm going to offer a slightly different take (but that's what makes the forums great, right?)

It appears as though you lost signal for one minute and two seconds, from 7m 32.8s to 8m 34s.  What I thought would have happened is that after 3 seconds of loss of signal the craft would have started an RTH.  However what happened with THIS flight appears to be that in the 1 minute of no RC signal, the craft continued to fly an additional 440 feet distance from home point.  I find this unusual and unexpected.

Upon recovery of the signal, the craft either enters RTH or was already in RTH.  This appears normal and expected.

At 9m 16s the craft determines it is critically low and (as expected and designed) it lands in place.  It has decided it does not have sufficient power to execute a full RTH.

Here is what I think COULD have happened next...  the application on the phone would have (should have) alerted you to this "auto landing" maneuver, and included in that warning is an option for you to abort the landing and to continue to keep control of the craft.  Do you recall if you saw such a warning and had such an option?  I don't know if you had enough battery and time to successfully RTH, but 18% is not insignificant, and it's possible that you could have navigated back.

Of particular interest is that one minute gap where it appears to me that you lost signal, and I would have thought it would start an RTH after 3 seconds of no signal.  That's the part that is of greatest interest, where the behavior differs from what I am use to and what I would have expected.

Some additional observations on your flight.  It appears that while you were in full control of the craft, you flew across a major road and over the parking garage.

I encourage you to review FAA regulations in regards to the road.  I won't critique this choice, but FAA does have some input on this.

The parking garage:  That is probably the cause of the problems between your RC (or phone) and the craft.  Parking garages contain a lot of metal and rebar.  This will interfere with the magnetics of the drone (particularly on liftoff), and will interfere with communications between craft and controller.  It is definitely not a desired situation to put a parking garage between you and your bird.

I would have still anticipated that the craft would have initiated an RTH after 3 seconds of disconnect.  I have no explanation as to why it didn't.

If you look at Time 8m 34s, you will see that Spark has initiated RTH itself because of connection loss (range limit reached in CE mode at 5,8 Ghz). Connection was later restored and maybe you interfered in flight direction. That part I cannot see from the logfile.  But if you read both logfile readings in parallel, you have had sufficient warnings beforehand to rather iniate RTH yourself at an earlier stage.
And to answer your other question: Yes the Spark has landed at the place of the tree, it had to land because of low battery. But you are somewhat lucky. FRom the Google Earth picture it is a good place with high chances to retrieve your Spark (worst case some tree climbing or high ladder involved ;-)

Hi Charles,
Thanks for your deep and detailed analysis. Great help!
In the meantime I've also spent some more minutes with the protocol and I've observed the same. As you've already mentioned: The slot where the drone flew for more than one minute just straight on is extremely odd. Also the way of the flight indicates that there might gone something wrong, since it is really real straight forward.
All in all this time interval and behavior is probably the main reason for my lost

As you've already said: After three seconds without signal my Spark should have begun with returning home... instead it flew away

Regarding your question... I can not remember that I've seen this a prompt with the option  to abort the landing.

schlomm Posted at 2017-9-11 13:10
Hi Charles,
Thanks for your deep and detailed analysis. Great help!
In the meantime I've also spent some more minutes with the protocol and I've observed the same. As you've already mentioned: The slot where the drone flew for more than one minute just straight on is extremely odd. Also the way of the flight indicates that there might gone something wrong, since it is really real straight forward.

That long straight line is not all by itself unusual.  What this means is that the log had one known location, and 1 minute later it had another known location, and it had no record of flight in between.  So it presumes a straight line, as that's the best it can derive from the available information.

Wow... you are fast in replying ;)
Yeah... I flew over a street, but this was definitely not my plan As far as I can remember, it was already impossible to control the Spark although the logfiles might show something different. At this. point I've already startet to run in the Spark's direction to get the control back.

As you can see at the beginning of my flight, I just spend some time over the large open space and I've never planned to leave this area (I am aware of the regulations).

Hopefully I'll find my Spark tomorrow... Ive spent two hours this evening, but it was to late and dark

Dear Charles and Schlomm, please accept my apologies when disagreeing. The first connection loss for more than 3 seconds occured at time 08m 34s and Spark correctly initiated RTH. I can fully understand that it reduces pain especially for a beginner if the fault can be found elsewhere, but one has also to accept that beginners faults are normal and can occur. This case remains pretty clear for me and cannot be seen as a fly away.

Wachtberger Posted at 2017-9-11 13:26
Dear Charles and Schlomm, please accept my apologies when disagreeing. The first connection loss for more than 3 seconds occured at time 08m 34s and Spark correctly initiated RTH. I can fully understand that it reduces pain especially for a beginner if the fault can be found elsewhere, but one has also to accept that beginners faults are normal and can occur. This case remains pretty clear for me and cannot be seen as a fly away.

No qualms about disagreeing at all.  And no apology necessary.  If we all had the same conclusions, we'd all be "you" or "me", and how boring would that be?  Our different interpretations help me learn how to read, understand and interpret these logs.

So to continue a conversation on different interpretations, I'm interpreting the gap from 7m 32.8s to 8m34s as a loss of signal.  The reason I draw that conclusion is because of the message at 8m 34s "Downlink Restored (after 1m 1.2s)".  The app appears to believe it suffered a 1m loss of comm.

And just to add without intention of increasing the pain. One may ask if for a beginner it was very wise to fly in sport mode despite three "Lage wind velocity" warnings before the problems started. I have at leat a few more "miles" with my Spark but still consider myself as absolute beginner and therefore never went into sport mode so far. There is a learning curve needed that should be respected ot otherwise potential consequences accepted.

Wachtberger Posted at 2017-9-11 13:40
And just to add without intention of increasing the pain. One may ask if for a beginner it was very wise to fly in sport mode despite three "Lage wind velocity" warnings before the problems started. I have at leat a few more "miles" with my Spark but still consider myself as absolute beginner and therefore never went into sport mode so far. There is a learning curve needed that should be respected ot otherwise potential consequences accepted.

There are a variety of flight decisions that were made that I would not have made myself.  This particular pilot seems receptive to honest and constructive critique (and thank you for that).

Charles Adams Posted at 2017-9-11 13:42
There are a variety of flight decisions that were made that I would not have made myself.  This particular pilot seems receptive to honest and constructive critique (and thank you for that).

Thank you Charles and I really admire your patience combined with deep knowledge, common sense and wisdom. Honestly you are a great asset for this forum. I have already learned a lot from you!

Wachtberger Posted at 2017-9-11 13:51
Thank you Charles and I really admire your patience combined with deep knowledge, common sense and wisdom. Honestly you are a great asset for this forum. I have already learned a lot from you!

Oh I can't tell you how many posts I've crafted, and then erased.  Patience, knowledge and common sense are a façade compared to what's going on really inside my brain!  

I see well what you are referring to and I am happy to let you deal further with the latest case from a few minutes ago... Being a US citizen you are also better placed and familiar with what is usual there.

Charles Adams Posted at 2017-9-11 13:57
Oh I can't tell you how many posts I've crafted, and then erased.  Patience, knowledge and common sense are a façade compared to what's going on really inside my brain!

Good to have that filter built in

Hello again,
Thanks again for your input!

As already said... no need to apologize! I'm absolutely fine with your postings and I'm very thankful for your opinions
Regarding my Flying-skills: I'm not a beginner and have some hours on my clock (indeed not with the Spark but with some others).
It is also not the case that I search the fault in DJI's responsibilities. Looking at the Flight protocol there are indicators where I should have been cleverer (as you already mentioned ;-)). Nevertheless I only can recapitulate the Spark's behavior and as far as I can remember I was not able to control the little cutie behind the ertain point. It's a pity that the RTH function did not work as I or we have expected. This is the only fact, where a big question mark is still in my mind.

Now I hope that I'll have uck tomorrow morning. I'll definitely keep you updated!

Thanks again for your honest posts!

schlomm Posted at 2017-9-11 14:09
Hello again,
Thanks again for your input!

And I honestly wish you good luck tomorrow! Seeing the location chances for retrieval are pretty good I believe. Just in case it is high up in the tree you maybe have a friend with copter who can help searching it from the air.

Thanks for your kind words
Unfortunately I do not know anyone who could help me with another drone, so I hope it works without. If not, I search for a new friend

Just one last question: Is it ensured that the drone did not flight in other other direction than downside since it was in the Auto Landing Mode?
As the last position is exactly over the right tree while the drone is in Auto Landing Mode I would concentrate my search on the right tree instead of checking the near roofs and other trees...

As for recovery I would try to start searching in the yard behind the
Institute für Erziehungswissenschaft, in particular by those trees. As for the logs they are indeed quiet strange. Did spark decide to fly straight across the street and the garage or was it you exploring the neighbourhood?

Schlomm, did you contact the support? Please provide me your ticket number or case number. I'll help you transfer the data. If not, please contact our support through this link: http://www.dji.com/support.
In addition,  http://www.phantomhelp.com/LogViewer is a third party viewer which we do not support. Uploading to dropbox and provide them to our support or post here. Then we can help you analyze the data.

Hi guys, also, appreciate if you guys can tell why the voltage seem to be in different color? Also the voltage seem to be low, then go back up?

I would guess on wind speed. I live near Münster and was flying my spark yesterday with extreme caution (on a field, max 30 m high). We had wind speed around 30 km/h yesterday, coming in gusts of more then 50km/h. Flying 100 meter high with the wind we had yesterday in the region was not well considered. Flying in Sport Mode around 50 km/h might be enough to fight against the wind, but when the signal was lost in RTH the drone should have gone to RTH speed around 30km/h. Maybe the reason for the 1 minute drift getting caught by strong wind.

If you get the drone back I would suggest not to fly in this conditions at great high with unpredictable wind speeds I would go that far to not even use it in town under this conditions but that's a personal decision to make.

Sorry for your loss, your log ends with 26m height so the drone maybe hit the roof of the building next to the trees? Just in case you can not find it in the tree.

Hello all,
Sorry for my late resonse, but I was quite busy during the last days.
Nevertheless a short update regarding my story: I got my spark back. It was found on the roof on one of the University buildings. Some really nice and helpful guys (probably the caretakers) helped me alot and after three days my spark was back
I already had some flight with my spark and everything is working! Now I'm only flying with a 50m long leash

Thanks for your support!