I flew my P4 in winter condition this morning where the temperature at the home point was -10C.  For the first 7 min, I basically did vertical from 0 to 60 meters and everything appeared to be normal. Once I started to fly forward, the camera immediately pointed at one of the motor pods and stuck there. The P4 flew backward in a circle for ~1.5km. I did not respond to my control. Then I got a warning message on the Go app - Warning: [compasss] data exception (in flight) switch to atti mode if craft behave abnormally (using)

I activated Return-to-Home and the P4 did not appear to respond. On second attempt, I can see on the map that the P4 began to fly in a straight line toward home point.  Soon after that, it returned to home point and landed safely.

I was aware of the temperature operation range of the P4 being 0 to 40C. It appears most comments posted here and vlogs in YouTube were with the battery. So I kept my battery warm until i was all set up to go. In retrospect, it was not a good idea given my SD card is not even rated to operate below 0.

Dear DJI representatives, is the compass data exception related to cold temperature? Should I put away all hope to capture winter sceneries?

I have uploaded the log:

http://healthydrones.com/main?share=nsTxex

The thing that comes in my mind is, you experienced a compass error, that's the good thing about our phantom's, they got a manual ATTI mode switch, so when the you noticed the error you should never use return to home, but you needed to put it in manual atti and bring it in manually. you were lucky that it worked

It was possible for me to fly it manually in ATTI mode now that I had time to review what happened.  Everything was happening very quickly at the time.  On top of that, the camera was stuck pointing at one of the pod.   I have uploaded below an example of the camera getting stuck which happened 10 days earlier in a much more manageable situation being 5 meters away from me.  Please see link below.  Again, this is not the footage of the camera issue when I had the compass data exception error.  But the symptom is exactly the same.



I speculate this problem was caused by a gust of wind that caught the camera and the gimbal was not able to compensate.
At this time I am grounding the P4 if the temperature is below 0C or if there is possibility of gusting wind.  Sadly, I have to write off filming  winter scenes.

Phantom Robin Posted at 2017-1-21 09:42
The thing that comes in my mind is, you experienced a compass error, that's the good thing about our phantom's, they got a manual ATTI mode switch, so when the you noticed the error you should never use return to home, but you needed to put it in manual atti and bring it in manually. you were lucky that it worked

is there a way to automatically enable flight logging on the card? i done see any other files than the vids on my card
my card only has a dcim/100media folder structure.

jsflysrc Posted at 2017-1-21 10:47
It was possible for me to fly it manually in ATTI mode now that I had time to review what happened.  Everything was happening very quickly at the time.  On top of that, the camera was stuck pointing at one of the pod.   I have uploaded below an example of the camera getting stuck which happened 10 days earlier in a much more manageable situation being 5 meters away from me.  Please see link below.  Again, this is not the footage of the camera issue when I had the compass data exception error.  But the symptom is exactly the same.

Did you calibrate the compass prior to take off ?

A couple of things stand out in the Healthy Drones report, one is the compass error, the other is that it reports a major deviation in voltage on cell 4 in the battery.

It is possible that there may have been a bad compass calibration, which is easily rectified. The battery seems a bit odd, as far as I can see it was up to temperature with your first 7 minute warming it nicely to 29C, unless there is a problem with it.

A number of others have and do fly regularly in low temperatures without problems so to fly at -10 shouldn't really cause any issues.

3 rd winter with phantoms
never had an issue with temps and flying
as long as battery is warm to start
-10c  is pretty normal here 10-30F in winter

good luck and have fun flying

WindSoul Posted at 2017-1-21 11:10
is there a way to automatically enable flight logging on the card? i done see any other files than the vids on my card
my card only has a dcim/100media folder structure.

flight data is stored on the device

not the sd card used for videos etc

Kirk2579 Posted at 2017-1-21 11:36
flight data is stored on the device

not the sd card used for videos etc

appreciated.

Aardvark Posted at 2017-1-21 11:25
Did you calibrate the compass prior to take off ?

A couple of things stand out in the Healthy Drones report, one is the compass error, the other is that it reports a major deviation in voltage on cell 4 in the battery.

No. I did not calibrate the compass before I took off. But the checklist in the GO app showed it was normal. But good point.  I should calibrate each time I go out.

jsflysrc Posted at 2017-1-21 10:47
It was possible for me to fly it manually in ATTI mode now that I had time to review what happened.  Everything was happening very quickly at the time.  On top of that, the camera was stuck pointing at one of the pod.   I have uploaded below an example of the camera getting stuck which happened 10 days earlier in a much more manageable situation being 5 meters away from me.  Please see link below.  Again, this is not the footage of the camera issue when I had the compass data exception error.  But the symptom is exactly the same.

i dont think you you have to completely give up winter scenes. my guess is your camera couldnt figure out the horizon line, hence the horizontal position. and now the long story:
something very similar happened to me today while flying in foggy conditions.
while over water, the cam would be adjusted properly. but when i started to fly along the water edge, the gimbals went on strike, like yours.
but then i turned towards water once more and the camera came back to normal (position).
my conclusion being that ambient light may play a determinant role in gimbals adjustment.

if someone asked me before, i would have said that gimbals work with drone to maintain camera orientation. but after today, i believe the gimbals work with the camera image to determine the horizon line.

now would be great if there was a setting to adjust gimbals on drone attitude (position in flight) alone. if someone knows of such a setting, the input will be appreciated.

jsflysrc Posted at 2017-1-21 11:40
No. I did not calibrate the compass before I took off. But the checklist in the GO app showed it was normal. But good point.  I should calibrate each time I go out.

no if it shows normal it is fine, do not re calibrate

he was asking to see if you did this time because that is often the cause of compass issues

good luck and have fun flying!

jsflysrc Posted at 2017-1-21 11:40
No. I did not calibrate the compass before I took off. But the checklist in the GO app showed it was normal. But good point.  I should calibrate each time I go out.

If it's got a good calibration then there should be no need to calibrate each time, that creates a higher risk of having a bad calibration (which is what I thought it may have been).

Something must have caused all those errors, you say both compasses are showing OK in DJI Go. I would  be tempted to try a firmware refresh & factory reset on the aircraft (using Assistant 2).

Edit:- Just watched your video, what sort of wind speeds were there ? I have never seen a gimbal stuck quite like that.

Kirk2579 Posted at 2017-1-21 11:48
no if it shows normal it is fine, do not re calibrate

he was asking to see if you did this time because that is often the cause of compass issues

the accelerometers provide the horizontal. the gimbals should receive at least a rough horizontal correction from the drone (camera sits on vibration dumpers), while the image processing should only fine tune it (instead of doing all work). sometimes (on slopes) the image gives no horizontal info, therefore the cam should stabilize with drone feedback alone, rather than give up on finding the horizontal. DJI, your input is appreciated.

Aardvark Posted at 2017-1-21 12:03
If it's got a good calibration then there should be no need to calibrate each tie, that creates a higher risk of having a bad calibration (which is what I thought it may have been).

Something must have caused all those errors, you say both compasses are showing OK in DJI Go. I would  be tempted to try a firmware refresh & factory reset on the aircraft (using Assistant 2).

From my experience in building electronic product and firmware programming, unless the components are rated and tested in extreme temperature (industrial temp -30C to 80C), there may be risk in output error.  (Example of digital compass that support industrial temp: https://www.sparkfun.com/datasheets/Sensors/Magneto/HMC5843.pdf).  According to the P4 specification, the operating temperature is rated for 0C to 40C.

Reprogramming the firmware may not correct the problem unless we speculate there is flash corruption.  If that is the case, the symptom most likely be more serious (like hard disk errors in your PC/blue screen of death which I do fear may happen one day while in flight).

Regarding to the gimbal,  you can see the there were some movement in the trees.  I estimate to be around 20km/hr gust pockets but otherwise calm.

WindSoul Posted at 2017-1-21 12:26
the accelerometers provide the horizontal. the gimbals should receive at least a rough horizontal correction from the drone (camera sits on vibration dumpers), while the image processing should only fine tune it (instead of doing all work). sometimes (on slopes) the image gives no horizontal info, therefore the cam should stabilize with drone feedback alone, rather than give up on finding the horizontal. DJI, your input is appreciated.

Please don't confuse people with your guesses. The image processing plays no part in the gimbal alignment.

jsflysrc Posted at 2017-1-21 09:11
I have uploaded the log:

http://healthydrones.com/main?share=nsTxex

Unfortunately Healthydrones only gives a basic summary and is next to useless to investigate flights incidents.
It's telling us that your problems were compass related but not much more.
There are a lot of guesses in this thread that aren't useful either.
To see actual flight data that is helpful go to http://www.phantomhelp.com/LogViewer/Upload/
Follow the instructions to upload your flight record.
Come back and post a link to the report it provides and we'll probably be able to get a better idea of what really happened.

Geebax Posted at 2017-1-21 12:47
Please don't confuse people with your guesses. The image processing plays no part in the gimbal alignment.

he way gimbals work suggests the horizontal is taken from the image.
reason being in this thread is posted a video where slanted lines made the camera go belly-up. also reason being my telling of a similar incident where again fog and absence of sun appears to smitten the camera.
the intuition says the horizontal should come from the drone. i call it intuition cause i didnt make the drone so cant be 100% positive on what the technical solution is in this case.
i hope this clears the air.
please make sure that you read the entire message and try to make sense before jumping to personal remarks. if you still have difficulties understanding, i am more than willing to explaining it at your level, so no need to turn confusion into ruffling feathers.

and since we are exchanging amabilities instead of focussing on solving threads, let me address the character posting next: i dont see the slightest resemblance of an insult in the message you replied off; are we still answering the thread?

WindSoul Posted at 2017-1-21 14:20
he way gimbals work suggests the horizontal is taken from the image.
reason being in this thread is posted a video where slanted lines made the camera go belly-up. also reason being my telling of a similar incident where again fog and absence of sun appears to smitten the camera.
the intuition says the horizontal should come from the drone. i call it intuition cause i didnt make the drone so cant be 100% positive on what the technical solution is in this case.

You have no idea how the gimbal works, and it does not help anyone here when you post your strange ideas. I have no trouble understanding what you wrote, but it is BS and has nothing to do with the issue. Therefore I asked you not to confuse people any further.

There was nothing personal in my response to you.

Geebax Posted at 2017-1-21 14:56
You have no idea how the gimbal works, and it does not help anyone here when you post your strange ideas. I have no trouble understanding what you wrote, but it is BS and has nothing to do with the issue. Therefore I asked you not to confuse people any further.

There was nothing personal in my response to you.

can you for the love of God explain it to a fellow who obviously has no idea? rather than BS yourself like a two-year old?can you for once stop calling yourself anyone and everyone and people? and talk (no more) in other people's name?
you know something? i give up on you. no matter whay your future contributions may be, ill treat them as they were never there.

WindSoul Posted at 2017-1-21 15:00
can you for the love of God explain it to a fellow who obviously has no idea? rather than BS yourself like a two-year old?can you for once stop calling yourself anyone and everyone and people? and talk (no more) in other people's name?
you know something? i give up on you. no matter whay your future contributions may be, ill treat them as they were never there.

Not sure what any of that means but I'll help you with this ...
"the way gimbals work suggests the horizontal is taken from the image"
The gimbal has no input from any image processing (if it did we'd always have perfect horizons).
It gets it's input from the sensors in the IMU inside the Phantom.
This is what tells the gimbal uses to understand what level is.

Labroides Posted at 2017-1-21 15:14
Not sure what any of that means but I'll help you with this ...
"the way gimbals work suggests the horizontal is taken from the image"
The gimbal has no input from any image processing (if it did we'd always have perfect horizons).

appreciated

Labroides Posted at 2017-1-21 13:42
Unfortunately Healthydrones only gives a basic summary and is next to useless to investigate flights incidents.
It's telling us that your problems were compass related but not much more.
There are a lot of guesses in this thread that aren't useful either.

Here is the log:
http://www.phantomhelp.com/LogViewer/8VMV5PUH37FL4EE7P43Z/

AlecW Posted at 2017-1-21 15:12
Why would anyone take time to even try to educate you when you use words like "BS yourself like a 2 year old" or "explaining it at your level"?  The best anyone can do is point out to others that your post is garbage advice and irrelevant.

Not only that... He kept whining and to the moderator from his last post about hitting the wall and he wants to moderate his own thread  


Protect animal population. Spay or Neutered and License your Pet!

jsflysrc Posted at 2017-1-21 15:46
Here is the log:
http://www.phantomhelp.com/LogViewer/8VMV5PUH37FL4EE7P43Z/

i'm no expert, but i think this is what happened.

Phantom 4 has a dual compass, designed to check for abnormalities.
I think it saw a deviation in the compass reading, reason is unknown to me but as designed the redundancy helped, and that's why the return to home worked as intented.
I believe phantom's are like computers, at anytime always something could go wrong, that's why i'm a advocate about a manual atti function switch, one should always be able to take manual control, thats a pro over the mavic, no manual atti switch there. I could advise you to a compas calibration, fly again in a control situation, see if the error comes back, if not i think it was a one time glitch, if it persists i believe you need to contact DJI to send it in for evaluation.

jsflysrc Posted at 2017-1-21 10:47
It was possible for me to fly it manually in ATTI mode now that I had time to review what happened.  Everything was happening very quickly at the time.  On top of that, the camera was stuck pointing at one of the pod.   I have uploaded below an example of the camera getting stuck which happened 10 days earlier in a much more manageable situation being 5 meters away from me.  Please see link below.  Again, this is not the footage of the camera issue when I had the compass data exception error.  But the symptom is exactly the same.

https://www.youtube.com/watch?v=XsuAWYEisu4&feature=youtu.be

I regularly fly my Phantom in sub zero temperatures (the coldest being -13 C) and have never encountered any issues like the one you describe. The Phantoms are perfectly capable of flying during winter, but there might be some side effects to consider - one is that the battery life is much shorter. I regard anything below 60% as EMPTY. Also make sure never to bring in the drone from the cold straight into a warm room, as this will cause condensation and may lead to moisture related problems. You need to be able to warm up the drone gradually.

I used it this winter at -20 degrees, light snowing and some wind. No problem just 2/3 of baterry (or less) available. The only problems my frozen hands (can't use it with gloves) and melting snow on my Iphone display.

It might help all of us if the promotions in status here @ the forum (beginner, hobbyist etc.) was also linked to total distant and time in air.  All that is needed now to advance in rank it seems is how  many posts and reply's
a person makes.
I think this is what leads to the creation of the DJI Jesus. (can't wait for his/her book to be published )  
I am listed as Hobbyist and I know I'm still a beginner at the drone world. In spite of my experience with electric and nitro helicopters, Transmitters. Gyro etc I know little about the phantom series
drones.
I'll give my 2 cents if there's a subject II know something about like Lipo batteries and interference and stuff like that, otherwise I just seek advice.
It would make me feel better if I knew the rank of the person responding to my post has earned the rank in the air and not at a keyboard. {:4_177:}
No one likes a dog that barks at everyone who walks by the yard.

So.... I have figured out who to take advice from. It's easy to tell, they never have to resort to childish negative responses, and they don't mind answering the same question over and over
no matter how many times it's been asked or at least post the thread  about this topic.  How about it MOD's is this asking too for too much ?

jsflysrc Posted at 2017-1-21 10:47
It was possible for me to fly it manually in ATTI mode now that I had time to review what happened.  Everything was happening very quickly at the time.  On top of that, the camera was stuck pointing at one of the pod.   I have uploaded below an example of the camera getting stuck which happened 10 days earlier in a much more manageable situation being 5 meters away from me.  Please see link below.  Again, this is not the footage of the camera issue when I had the compass data exception error.  But the symptom is exactly the same.

https://www.youtube.com/watch?v=XsuAWYEisu4&feature=youtu.be

Few weeks ago, same problem like yours, gimbal stucked on weird position.
Temperature was -10 degree and some wind, over frozen Danube river.
Had some message on IPhone something about gimbal has reach the maximum .... didn’t remember unfortunately.
In about 10 sec everything back to normal. After 30 min (new battery) same issue but just for few sec.

Till then, use it in warmer temp (Myanmar over 30 degree) for few hour and no issue.

Maybe the temp combined with heavy wind is the problem.

jsflysrc Posted at 2017-1-21 11:40
No. I did not calibrate the compass before I took off. But the checklist in the GO app showed it was normal. But good point.  I should calibrate each time I go out.

Just calibrate once.   I bought my drone  a month ago.  Calibrated compass when started preparing for flight.  Then again after a firmware update and also IMU calibration. Have done around 60 flights since firmware update and never needed to do another compass calibration. P3S drone.

WindSoul Posted at 2017-1-21 15:00
can you for the love of God explain it to a fellow who obviously has no idea? rather than BS yourself like a two-year old?can you for once stop calling yourself anyone and everyone and people? and talk (no more) in other people's name?
you know something? i give up on you. no matter whay your future contributions may be, ill treat them as they were never there.

Windsoul you have had many times the same issue that you insult people all the time because they correct your mistakes.  Then you want them to explain it to you like you are a 2 year old. Now that we know you think like a 2 year old I have reply with this.  Mother nature was obviously quite restrictive with the intelligence you were given. Drones are for adults or people who at least think as adults not 2 year olds. I suggest you pack yours up and go elsewhere because in nearly every thread you post your information is nonsense.  

You need to read the manual about cold weather conditions which can affect the operation of your drone and especially the batteries which need a minimum operating temperature.

Below -10C would be considered extreme cold. I do fly up where I live where my drone can encounter sub zero temperatures and if you add in wind chill factor you can see that the speed of flight could also be an issue in extreme cold.

There was also a post about how the gimble can stick on the cable behind if there is excess glue on the cable.  So the cable behind the gimble needs to be looked at.

For what it's worth, I've flown in temperatures as low as -34C.   I have dozens of flights where the ambient air temps were below -18C (0 degrees F).   (We had a really cold December ).

I've had no issues with control.  Not only that, but I haven't seen any reduction in flight times.