First time this has happened. I did everything by the book (or so I think). Open field. No apparent magnetic interferance or IMU issues. Checked before flight and even once during flight via DJI app. All clear. Connected at first via wifi to the remote, virtually no interferance in the area as shown in app, then connected via OTG cable. Before takeoff I ensured that the app and the AC were pointing in the same and correct direction. 17 satelites and 5 bars throughout the flights. Until, and without any previous warning, I get messages about Yaw Error and "In Flight, working IMU encounters heading exception,please switch to atti mode if craft behave abnormally; Yaw Error" and "Weak GPS". Luckily I was low to the ground. Flipped into sport and very very carefully flew it back to me and ended the flight. What on earth could this have been??
Some things of note
I notice that in general the AC tends to rotate slowly sometimes when hovering. It's always done this.
Often the app compass doesn't point where I know I'm pointing, and I have to move the phone in figure of 8 pattern. This particular flight it seemed to take a bit longer. This is before the AC was turned on.
The last thing I noticed, and in hindsight should have caused me to end the flight, is that the AC started to deviate it's straight ahead course by about 10-15 degrees. I thought maybe it was the wind pushing it. Shortly after I got the errors.

http://www.phantomhelp.com/LogViewer/OQHXLJQJJ1BNK9AA5YY1/

You took off from what looked like small road or dirt track, could you have picked up some interference from there?
Regarding aircraft veering off flying at such low altitude over that type of ground this can happen if ground is not textured enough then VPs can have trouble getting good lock on the ground.

IMU the exception heading is nothing to do with bad IMU it’s a consequence of compass problem, however you did say Aircraft was involuntarily yawing this can be a problem with IMU as it is the job of Imu to maintain the attitude of your aircraft, so doing an IMU calibration won’t do any harm.

Triangle if this is off you could try calibration on your phone to see if this corrects this remember just like Aircraft no metal objects etc etc.

hallmark007 Posted at 2017-12-9 13:31
You took off from what looked like small road or dirt track, could you have picked up some interference from there?
Regarding aircraft veering off flying at such low altitude over that type of ground this can happen if ground is not textured enough then VPs can have trouble getting good lock on the ground.

I took off from The very edge of a parking lot facing a wide open field. The surface was asphalt, and to my knowledge it doesn’t contain any rebar, but now that I think about it I was near to a curb made of concrete. I’m wondering if that could have any metal in it? Something else I should’ve mentioned, my car was parked about 25 feet behind me. However I checked the Compass reading in the app and looked OK, so I seemed I was far enough, especially because the manual doesn’t specify a distance to stay away from metal structures. At the moment I got the errors, I was sitting in the car. With the door open and facing the field. So, like half sitting in it. Could this be a factor? Lastly, the ground was dirt and grass. And the AC seems to track it fine for the first half of the flight, and then for whatever reason, it veered to the right by about 10°.The ground had sufficient detail and the lighting conditions were good. Admittedly I was only 12 feet from the ground. When calibrating the phone compass, I wore my wedding ring. Would that matter? Also, is it better to calibrate the phone with the remote attached, or by itself? Definitely going to give the I am you calibration a shot.

Other than interference, what with another possible cause of the yacht error I am you exception be? If there was interference, why do you suppose I wouldn’t see it in the app before or during flight? One last thing I’m also remembering. I downloaded an Eat and after sensor/magnetometer app. And I tested it at the take off point, and the reading was 55 micro teslas. Would that be high enough to cause issues? Just trying to learn as much as I can to prevent something like this from happening again. Was pretty freaky.

Really appreciate your experience, as well as anyone else who cares to chime in. Thank you.

ernie_halter Posted at 2017-12-9 15:31
I took off from The very edge of a parking lot facing a wide open field. The surface was asphalt, and to my knowledge it doesn’t contain any rebar, but now that I think about it I was near to a curb made of concrete. I’m wondering if that could have any metal in it? Something else I should’ve mentioned, my car was parked about 25 feet behind me. However I checked the Compass reading in the app and looked OK, so I seemed I was far enough, especially because the manual doesn’t specify a distance to stay away from metal structures. At the moment I got the errors, I was sitting in the car. With the door open and facing the field. So, like half sitting in it. Could this be a factor? Lastly, the ground was dirt and grass. And the AC seems to track it fine for the first half of the flight, and then for whatever reason, it veered to the right by about 10°.The ground had sufficient detail and the lighting conditions were good. Admittedly I was only 12 feet from the ground. When calibrating the phone compass, I wore my wedding ring. Would that matter? Also, is it better to calibrate the phone with the remote attached, or by itself? Definitely going to give the I am you calibration a shot.

Other than interference, what with another possible cause of the yacht error I am you exception be? If there was interference, why do you suppose I wouldn’t see it in the app before or during flight? One last thing I’m also remembering. I downloaded an Eat and after sensor/magnetometer app. And I tested it at the take off point, and the reading was 55 micro teslas. Would that be high enough to cause issues? Just trying to learn as much as I can to prevent something like this from happening again. Was pretty freaky.

First you could have got compass interference from the curb, obviously I can’t be certain but it’s possible.
25 metres is plenty distance to be from your car.

Second sitting in your car with remote is not good, your RC contains small magnets which can be effected by interference this also could be a cause for your AC veering 10 degrees.

Third why you didn’t get or see warnings, one of the reasons may well be you never lost gps you did say you got weak gps warning.

Forth I wouldn’t calibrate phone with RC, but yes no metal will help.

Fifth yaw error is caused by bad compass heading you are pushing to go one way and Aircraft compass is forcing Aircraft to go on its heading and IMU try’s to obey command from controller so is sending message back to controller that it can’t keep heading because bad compass, then next step would usually be Aircraft to drop gps in favour of Atti mode, this didn’t happen simply because interference was not so strong.

Sixth 55 micro Teslas sounds strong I am not aware of the app and I’m not sure or confident to give you an answer on this, perhaps others will be more qualified to give you a better answer.

Thank you. Correction I was 20-25 feet not meters. (Also sorry to all for the weird typos. Was dictating via Siri. )

Re the magnetometer. I’ve tried using several apps and they all give the same results in response to EMF. They could also all be wrong in the same way. I realize though that I tested it while holding it. I should have set it on the ground to test.

Still, it’s very puzzling and disconcerting to me that every sensor shows good conditions minutes before a potential flyaway. Am I missing something?

Is it possible that hand launching would have made a difference today.

Something else I’m remembering  — often, if I look at my position and heading on my phone, using Google apps or similar, the heading is as it should be. Then I check it in the DJI go app and it’s different. This is with no aircraft connected. Assuming I am able to resolve it by calibrating the iPhone compass, if I inadvertently lock/unlock the screen i’m back to square one. Is this normal, Unknown bug? Anyone else with this issue?

ernie_halter Posted at 2017-12-9 18:16
Something else I’m remembering  — often, if I look at my position and heading on my phone, using Google apps or similar, the heading is as it should be. Then I check it in the DJI go app and it’s different. This is with no aircraft connected. Assuming I am able to resolve it by calibrating the iPhone compass, if I inadvertently lock/unlock the screen i’m back to square one. Is this normal, Unknown bug? Anyone else with this issue?

Hi Ernie, sorry about replying your question with another question, but should you have a compass in you mobile? Is it a DJI requirement that your mobile must have built in compass?

ernie_halter Posted at 2017-12-9 17:41
Thank you. Correction I was 20-25 feet not meters. (Also sorry to all for the weird typos. Was dictating via Siri. )

Re the magnetometer. I’ve tried using several apps and they all give the same results in response to EMF. They could also all be wrong in the same way. I realize though that I tested it while holding it. I should have set it on the ground to test.

Sensors may show good , but I would always advise to allow spark to hover at around 2 metre close to make sure everything is ok. I have left preflight check below if it can be of any use.

If you hand launched , well it’s hard to give yo an answer to that one, I think you should go back to location and test another flight if you can, make sure you launch site is safe and take it from there.

https://forum.dji.com/forum.php? ... &fromuid=260008

It was false alarm.....this had happen to me 4 times. It was never recorded in flight records,  the aircraft did not drift or fly away or anything abnormal....I think it's app bug...

Look at my log here: http://phantomhelp.com/logviewer/TAY6Q46ROMRVOTQIAJFA/

Everything was fine and then all of a sudden it went rogue. I tried to control it but image transmission was so poor that I had no idea where it was heading. After some struggle, I managed to get it closer to me and then it just DCed. I was watching it drifting away but couldn't control it anymore.

Before this flight I had a similar experience which ended well, but this time I wasn't so lucky. I read all manuals and tutorials, watched all videos on how to prevent flyaways, but unfortunately, it wasn't enough.

djiuser_7DCDCMv Posted at 2017-12-10 05:23
Hi Ernie, sorry about replying your question with another question, but should you have a compass in you mobile? Is it a DJI requirement that your mobile must have built in compass?

If I understand you correctly, you're asking if having a compass in your mobile device / phone is necessary? If so, I was wondering the same thing.

Seeing as the DJI Go App runs on only iPhone IOS and Android, and any phone running those operating systems would have a compass / GPS in the phone I imagine.

I don't imagine your phone's compass is required for flight since the Spark can fly without a mobile device altogether. However,  I'm curious if a bad compass on your phone / DJI app would cause errors with the spark in flight?

Anyone know?

hallmark007 Posted at 2017-12-10 06:17
Sensors may show good , but I would always advise to allow spark to hover at around 2 metre close to make sure everything is ok. I have left preflight check below if it can be of any use.

If you hand launched , well it’s hard to give yo an answer to that one, I think you should go back to location and test another flight if you can, make sure you launch site is safe and take it from there.

Thank you, I remembered your flight check list from another post and did follow the hover, back 2 ft, forward 2 ft, sides etc. It seemed fine, with the exception of the slow yawing while hovering, which it's always done since I purchased about a month ago. Now thinking the slow yaw may be related to an overall compass problem, or at least more significant than I thought.

It's unlikely I'll be in that area again as I was traveling through and stopped to fly there. It's about 200 miles from where I live.  If I am near there though I'll be sure to check it out again and see what more I can learn.

So if everything is as it should be, with regards to IMU / compass, the Spark should hover and lock onto the same heading without any slow corrections or drift. Or is there an acceptable amount?

Just ordered a handheld analog compass to see if I can get to the bottom of this. : ]

Anuvis Posted at 2017-12-10 06:24
It was false alarm.....this had happen to me 4 times. It was never recorded in flight records,  the aircraft did not drift or fly away or anything abnormal....I think it's app bug...

Though you're right, mine didn't go into ATTI mode but it was flying straight when pushing forward on the sticks. Scary stuff.

choban Posted at 2017-12-10 06:44
Look at my log here: http://phantomhelp.com/logviewer/TAY6Q46ROMRVOTQIAJFA/

Everything was fine and then all of a sudden it went rogue. I tried to control it but image transmission was so poor that I had no idea where it was heading. After some struggle, I managed to get it closer to me and then it just DCed. I was watching it drifting away but couldn't control it anymore.

Ugh! Yea I saw that. So sorry man. Hopefully DJI replaces it under warranty. I did read what Hallmark said about your takeoff point being off of concrete, as my takeoff point was off asphalt but near a concrete curb. Wondering if that was the cause of the compass problems.

If you're in that same location, would you mind getting your iPhone compass app  open and set your phone in the exact place you took off from. See if the compass deviates in that spot?

ernie_halter Posted at 2017-12-10 09:41
Though you're right, mine didn't go into ATTI mode but it was flying straight when pushing forward on the sticks. Scary stuff.

meant to say it *was NOT* flying straight. fingers moving faster than my brain at the moment.

ernie_halter Posted at 2017-12-10 09:44
Ugh! Yea I saw that. So sorry man. Hopefully DJI replaces it under warranty. I did read what Hallmark said about your takeoff point being off of concrete, as my takeoff point was off asphalt but near a concrete curb. Wondering if that was the cause of the compass problems.

If you're in that same location, would you mind getting your iPhone compass app  open and set your phone in the exact place you took off from. See if the compass deviates in that spot?

We'll see...waiting for their response. I took off that time from a gravel path, but not sure if there was any concrete underneath.

Good point, I will definitely check it with my phone's compass.

Btw. when I took off I checked if the arrow on the map was pointing in the same way as Spark did, and that was fine. It's tough now to know exactly what's causing these errors...every case is different. Maybe DJI can find out more from .DAT files.

ernie_halter Posted at 2017-12-10 09:44
Ugh! Yea I saw that. So sorry man. Hopefully DJI replaces it under warranty. I did read what Hallmark said about your takeoff point being off of concrete, as my takeoff point was off asphalt but near a concrete curb. Wondering if that was the cause of the compass problems.

If you're in that same location, would you mind getting your iPhone compass app  open and set your phone in the exact place you took off from. See if the compass deviates in that spot?

With a good IMU it should on a windless day hold everything very steady once you have gps lock and good lighting conditions.

Some tips about IMU.

IMU "Inertial measurement unit" .

As far as I see it, IMU calibration on a level surface updates a table of values the flight controller software uses as a reference for a level stationary hover. From there the craft responds predictably to flight commands.  It also likely measures any sensor noise and thermal drift so that these technical imperfections are accounted for when using IMU sensor data in flight. - this is likely why there must be no vibrations during the calibration process.


Bad IMU calibration could cause drift and attitude issues as the flight controller fights to hold the craft in what it thinks is the correct attitude as opposed to the correct physical attitude.


The IMU usually has 2 types of sensors – angle and acceleration and in turn 3 sensors of each type measuring in the X,Y and Z axis. These sensors can, through vibration, aging, impact etc, drift in their response over time and thus an IMU calibration will establish a new reference for the IMU’s level/stationary state that the flight controller can work with to restore stable flight.


Think of calibration as the bringing back into line the measured craft attitude with the true physical craft attitude.


As to how often an IMU calibration is needed, Refer to your manual ; certainly after any impact or if there are suspicions the craft can’t hover in a level attitude in a windless environment.

hallmark007 Posted at 2017-12-10 10:11
With a good IMU it should on a windless day hold everything very steady once you have gps lock and good lighting conditions.

Some tips about IMU.

Thank you! I'm a little familiar with IMU's since I've built quads with various CC3D flight controllers and set PIDS and all that good stuff. It seems to hover ok, minus the slow yawing at times. But tested it in my front lawn just not and got another compass/IMU set of errors. Just before the errors, I confirmed that pushing forward on the sticks did not result in a straight ahead motion but slightly deviated 10-20 degrees. Compass readings in the app were good. Solid GPS.

Checked the iPhone compass, calibrated it, and pointed myself towards true north. then pointed the spark on the grass in the same direction. Connected the remote and the AC and suddenly the app thinks I'm pointing about 30 degrees to the left. I removed the phone from the remote and recalibrate the phone compass. They align, I carefully test fly simple motions, and then of course I ended the flight immediately after the errors.

I had the AC on the box, relatively level, and never touch or move it during the boot up sequence.

I really don't know for sure  everything *seems* to be suggesting compass issues vs IMU. I'm going to try deleting and re-installing the app as well as confirming headings with an analog compass.

How common is it to have the DJI Go app tell you you're pointing in a direction you know you aren't?  This seems to happen to me every single time. That can't be normal, am I correct?

Looking back on my logs from yesterday, same thing. The AC never went into ATTI mode but throughout the flight (especially after the errors), pushing straight on the sticks resulted in a non-straight ahead motion. Like off by about 30-45 degrees.

Thinking this is probably time to calibrate the compass. Going to wait until I am in a wide WIDE open area, no metal, and follow directions to the letter.

Decided against compass calibration. Instead have started to rely on an analog compass to verify magnetic north after I discovered the RC itself can cause magnetic interference from the magnets inside it.