How do I go about finding out the reason for my fly-away in the flight info?

I had my first fly-away last week. I calibrating my bird and it kept telling me I had magnetic interference so I moved 25-50 feet in places until I did not get the warning. I flew in pretty much the same spot a few days prior with no warning. My buddy was flying his P4 and was getting the same warning as I was.

Anyhow I got the p3s up and flew with no issues. However upon landing about 5-8 feet from the ground it took off violently to the left. I immediately brought it down but not before hitting a concrete wall, flipping over and crashing. Surprisingly no physical damage upon inspection, not even a bent or broken prop.

I took it flying yesterday for the first time since mishap and flew just fine. I was pretty relieved that there appears to me no lasting damage. I even risked using the same props. Everything checked out just fine.

I tell myself quadcopters are kind of like harddrives, its not "if they will fail", its "when they will fail"

I think most of us regard the phrase fly-away as a situation where your Phantom disappears over the horizon.  

Anyway, if you could put your csv flight log into this site and post the link here, we can look at it and maybe figure out what happened.

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

Mark The Droner Posted at 2017-3-5 06:22
I think most of us regard the phrase fly-away as a situation where your Phantom disappears over the horizon.  

Anyway, if you could put your csv flight log into this site and post the link here, we can look at it and maybe figure out what happened.

terminology, Thank you for the correction, I used it thinking a fly away was something it does that is out of my control.
How would you describe that?

I don't know.  Faultless-Mishap?  Manufacturer-Liable-Accident?  Act of God/Buddha?  

No biggy.

If you had wrecked it, I'd call it a Rapid Unscheduled Disassembly.  

Mark The Droner Posted at 2017-3-5 06:22
I think most of us regard the phrase fly-away as a situation where your Phantom disappears over the horizon.  

Anyway, if you could put your csv flight log into this site and post the link here, we can look at it and maybe figure out what happened.

link  
I am using an android and I think this is the correct path to up load the log?  phone/dji/dji-pilot/log/crash

http://www.phantomhelp.com/LogVi ... 2017-02-24-15-05-32[1].txt


java.lang.NullPointerException: Null pointer exception during instruction 'monitor-enter v1'
        at com.nokia.maps.PlacesCategoryGraph$a$1.a(Unknown Source)
        at com.nokia.maps.PlacesCategoryGraph$a$1.onCompleted(Unknown Source)
        at com.nokia.maps.PlacesBaseRequest$4.run(Unknown Source)
        at android.os.Handler.handleCallback(Handler.java:739)
        at android.os.Handler.dispatchMessage(Handler.java:95)
        at android.os.Looper.loop(Looper.java:145)
        at android.app.ActivityThread.main(ActivityThread.java:6872)
        at java.lang.reflect.Method.invoke(Native Method)
        at java.lang.reflect.Method.invoke(Method.java:372)
        at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:1404)
        at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:1199)

It should be a csv file even if it ends with txt.  In other words, it's a data log that would look like a spread sheet but the data is separated by commas.  I believe.  I confess I don't use Go so I can't tell you for sure.  

chainmon Posted at 2017-3-5 06:25
terminology, Thank you for the correction, I used it thinking a fly away was something it does that is out of my control.
How would you describe that?

I think I found it.
I know my battery was low, in fact I try and not fly anything drained less than 30% and this time it was about 20% when I came in to land. Yaw and speed error?

http://www.phantomhelp.com/LogViewer/KCZEG570FOT4FVSR0KGI/

chainmon Posted at 2017-3-5 06:58
I think I found it.
I know my battery was low, in fact I try and not fly anything drained less than 30% and this time it was about 20% when I came in to land. Yaw and speed error?

And I am not sure why it shows ATTI right about when I was having issues near the ground. I did not flip the P switch down  p3S
https://drive.google.com/file/d/ ... Wc/view?usp=sharing

I don't know.  Nothing obvious is jumping out at me.  Other than the yaw error of course.  

Found this thread which seems to be discussing a similar situation.   There is some optimism is Post 21.  

http://www.phantompilots.com/thr ... or-yaw-error.96675/

Mark The Droner Posted at 2017-3-5 07:29
I don't know.  Nothing obvious is jumping out at me.  Other than the yaw error of course.  

Found this thread which seems to be discussing a similar situation.   There is some optimism is Post 21.  

very odd...
I flew yesterday for the first time since mishap and no errors.

http://www.phantomhelp.com/LogViewer/TSZEPM7ZO4T3OBSQ91GI/

Mark The Droner Posted at 2017-3-5 06:22
I think most of us regard the phrase fly-away as a situation where your Phantom disappears over the horizon.  

Anyway, if you could put your csv flight log into this site and post the link here, we can look at it and maybe figure out what happened.

agreed on definition
term is over used..

BTW
from picture I would say there is a lot of rebar and steel substructure...

good luck and have fun flying!

chainmon Posted at 2017-3-5 06:25
terminology, Thank you for the correction, I used it thinking a fly away was something it does that is out of my control.
How would you describe that?

good question!

maybe break down the true "flyaway" as to a defined fault of the quad itself,
usually found in the log..

as oposed to something that occurs from:
operator error
mis use of an app feature
failure to understand RTH
failure to understand proper calibration procedures and needs.
panic
flying out of VLOS when not truly prepared..
taking off from area with lots of metal
etc, etc

good luck and have fun flying!

chainmon Posted at 2017-3-5 07:35
very odd...
I flew yesterday for the first time since mishap and no errors.

I would recommend syncing your flight logs and submitting them to Support in order to review them. From looking at what you've provided, I would agree with post #11 that you were most likely incurring interference from the surrounding metal. Try flying in a more open space and away from metal to see if this persists.

chainmon Posted at 2017-3-5 07:05
And I am not sure why it shows ATTI right about when I was having issues near the ground. I did not flip the P switch down  p3S
https://drive.google.com/file/d/0B8skTpCaa0xAMk0tLTB3YU9IaWc/view?usp=sharing

Your log for that time is laced wit yaw errors which are a kind of compass error.
When your Phantom has a compass error (or yaw error) it automatically goes to atti mode because of the conflicting data it's getting from compass and GPS.
There isn't enough evidence to be certail but in similar cases I've seen, the cause was launching from reinforced concrete surfaces which screws up your compass.
What can you say about where you launched from?

Everytime I have flown from areas of heavy concrete with rebar in I get the same compass error due to the magnetic field from the iron rebars. The drone will often behave erratically as it gets close to the ground and enters this rebar magnetic field. Best not to fly at all in these areas. If you see concrete anywhere its likely to have rebar in if it is load bearing. Useful lesson here for you about where and where not to fly.j I suspect most people in urban locations have come across this error at one point or another.

Kirk2579 Posted at 2017-3-5 08:34
good question!

maybe break down the true "flyaway" as to a defined fault of the quad itself,

yes this was the fault of the drone with a little user error flying near magnetic interference

Kirk2579 Posted at 2017-3-5 08:28
agreed on definition
term is over used..

Glass Museum, yet steel structure. I must have gotten lucky a few nights ago when I flew from this spot. Actually I flew twice from this spot the week before without incident.

Mark The Droner Posted at 2017-3-5 07:29
I don't know.  Nothing obvious is jumping out at me.  Other than the yaw error of course.  

Found this thread which seems to be discussing a similar situation.   There is some optimism is Post 21.  

Mark,
Thank you for the link

DJI-Jamie Posted at 2017-3-5 22:27
I would recommend syncing your flight logs and submitting them to Support in order to review them. From looking at what you've provided, I would agree with post #11 that you were most likely incurring interference from the surrounding metal. Try flying in a more open space and away from metal to see if this persists.

Jamie,
syncing and flight logs...more research to do,  Off googling

Labroides Posted at 2017-3-5 23:12
Your log for that time is laced wit yaw errors which are a kind of compass error.
When your Phantom has a compass error (or yaw error) it automatically goes to atti mode because of the conflicting data it's getting from compass and GPS.
There isn't enough evidence to be certail but in similar cases I've seen, the cause was launching from reinforced concrete surfaces which screws up your compass.

"When your Phantom has a compass error (or yaw error) it automatically goes to atti mode"
Labroides, Thank you, I did not know this.

Labroides Posted at 2017-3-5 23:12
Your log for that time is laced wit yaw errors which are a kind of compass error.
When your Phantom has a compass error (or yaw error) it automatically goes to atti mode because of the conflicting data it's getting from compass and GPS.
There isn't enough evidence to be certail but in similar cases I've seen, the cause was launching from reinforced concrete surfaces which screws up your compass.

Labroides,
Yes if you look at the image of the link
http://www.phantomhelp.com/LogViewer/KCZEG570FOT4FVSR0KGI/
there is a lot of rebar I am sure. I am launching from a big dock. I  won't be launching from here again.

Not to highjack the thread but, when you mentioned to the OP that flying near the concrete, steel, & rebar structure was bad due to the magnetic field surrounding the building, is this only for take-off or anytime during flight? I haven't heard of this being a large factor before? How do many pilots fly in big cities with nothing but buildings or around new contruction areas for surveying/photos?

TPVAerials Posted at 2017-3-6 06:05
Not to highjack the thread but, when you mentioned to the OP that flying near the concrete, steel, & rebar structure was bad due to the magnetic field surrounding the building, is this only for take-off or anytime during flight? I haven't heard of this being a large factor before? How do many pilots fly in big cities with nothing but buildings or around new contruction areas for surveying/photos?

Launching from or calibrating the compass on or close to reinforced concrete is the problem.
There have been many cases, it's not a rare thing.
Flying in the vicinity of city buildings isn't the problem - having your compass sitting right on top of steel is..

Labroides Posted at 2017-3-6 06:09
Launching from or calibrating the compass on or close to reinforced concrete is the problem.
There have been many cases, it's not a rare thing.
Flying in the vicinity of city buildings isn't the problem - having your compass sitting right on top of steel is..

Ok, thanks for the clarification. That was my original understanding, but seeing the answers above had me questioning that info.

I recently landed on a blacktop parking area; I shutdown for a few minutes and then attempted to startup and got the compass error message.  The Phantom 3 Standard would not start, I tried three times without success.  I then moved the Phantom 3S about ten feet to a new location, and it started as normal!  My conclusion is that there was some large metal hidden under the blacktop.

chainmon Posted at 2017-3-6 05:33
yes this was the fault of the drone with a little user error flying near magnetic interference

sorry but not the fault of drone
if it does as designed and within limits of onboard hardware.

in other words the user set it up incorrectly/improperly and this resulted.

had no compass calibration been done at that site , I doubt we would be reading this!

good luck and have fun flying..

these are ready to fly out of box
but the operator learning curve is very long and time consuming..

Kirk2579 Posted at 2017-3-6 11:23
sorry but not the fault of drone
if it does as designed and within limits of onboard hardware.

Kirk,
so you are saying "had no compass calibration been done at that site , I doubt we would be reading this!"

I should not have calibrated the p3s? DO I understand that correctly? Or could I have done it and I did not realize it did not take properly?
Plz advise

chainmon Posted at 2017-3-7 04:31
Kirk,
so you are saying "had no compass calibration been done at that site , I doubt we would be reading this!"

Did you recalibrate the compass at the site of this flight?

You don't need to recalibrate the compass before each flight and doing so makes you less safe - not more safe.
Get a good compass calibration in a clear area and stick with it.

Labroides Posted at 2017-3-7 06:58
Did you recalibrate the compass at the site of this flight?

You don't need to recalibrate the compass before each flight and doing so makes you less safe - not more safe.

So I calibrate every time I fly. There are times I fly 10-18 miles from where I flew the previous flight. So you are saying I should not do this? I should not calibrate every time? How often should this be done?

Yes, that's what he's saying.  Experienced pilots do not calibrate every flight.  The last time I calibrated, I couldn't even remember for sure how to do it - it had been so long.  I literally had to check the manual.  I'm not making this up.

Why not calibrate every flight?  Because you might end up with a bad calibration whereas you had a good one before you calibrated.  

Re distance, some say 50 miles, some say 100 miles, some say 300 miles distance before calibration is needed.  

Bottom line:  If you are reasonably sure you had a good calibration last flight, there is no need to calibrate next flight - unless you think you've traveled some great distance since the last flight.

I had a low altitude stabilization problem a few days ago, it happened first flight after updating DJI APP, P3 would not hoover and went left and right i was just barely able to land it without it tipping over. I flew a couple of days later and it worked fine.   

Mark The Droner Posted at 2017-3-7 09:36
Yes, that's what he's saying.  Experienced pilots do not calibrate every flight.  The last time I calibrated, I couldn't even remember for sure how to do it - it had been so long.  I literally had to check the manual.  I'm not making this up.

Why not calibrate every flight?  Because you might end up with a bad calibration whereas you had a good one before you calibrated.  

interesting,
This video must be wrong then?


He says calibrate at each new destination.

People have different opinions on how far one must travel to require a compass calibration.  In this video, it depends on how you define destination.  The guy in the vid seems to define it as different continents.  

Next time you fly, think about where you are relative to magnetic North.  

Then move some place else and consider flight.  Think about how much magnetic north has changed relative to east and west since the last flight.  It probably hasn't changed much.  

You would have to travel quite a distance for it to matter.  

Hope this helps.

I cannot launch within around 30m of this bridge due to magnetic interference.. I also have an Argtek antenna on my P3S.  So I also stay from it when taking this video. There are also power lines to take into consideration. I have several of these bridges near where I live.  I won't even try to fly under them where they span rivers.  Watch out for the birds as well.

chainmon Posted at 2017-3-7 11:19
interesting,
This video must be wrong then?

The whole calibrate-every-time thing is a superstitious ritual that was never necessary and caused quite a few crashes.
It was a myth caused by bad wording in the earlier DJI manuals.
Look at the manuals for the newer drones and they don't say anything about recalibrating at any distance or every flight or after updates.

Labroides Posted at 2017-3-7 13:42
The whole calibrate-every-time thing is a superstitious ritual that was never necessary and caused quite a few crashes.
It was caused by bad wording in the earlier DJI manuals.
Look at the manuals for the newer drones and they don't say anything about recalibrating at any distance or every flight.

I only calibrated after a firmware update but never since and have over 100 flights since update.

Mark The Droner Posted at 2017-3-7 09:36
Yes, that's what he's saying.  Experienced pilots do not calibrate every flight.  The last time I calibrated, I couldn't even remember for sure how to do it - it had been so long.  I literally had to check the manual.  I'm not making this up.

Why not calibrate every flight?  Because you might end up with a bad calibration whereas you had a good one before you calibrated.  

1 year this month for me on compass calibration....

chainmon Posted at 2017-3-7 09:24
So I calibrate every time I fly. There are times I fly 10-18 miles from where I flew the previous flight. So you are saying I should not do this? I should not calibrate every time? How often should this be done?

never unless you have an issue and the app says too.

a basic interference message is not telling you to calibrate
that usually mean you are too close to a metal object when starting up

chainmon Posted at 2017-3-7 11:19
interesting,
This video must be wrong then?

Wow, there are so many things wrong with that video. Please do not use it as an example. The first time he shows calibrating ther compass, he is standing next to a metal chair - NO! He is also standing on a wooden deck that may have a steel frame - NO!

He then says: 'The GPS module has a built in magnetic field sensor for measuring the geomagnectic(sic) field' . GPS and compass are not related, they are not even in the same place inside the aircraft. The GPS receiver is in the top of the main body and the compass is in the bottom of one of the legs.

All this video does is perpetuate the same old-wives tales about compass calibration.