DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

New FAA hobby regs (17th of May 2019) and DJI firmware update (or lack thereof)

Thanks... yea well I might have exaggerated a bit with "swan dived", but the aircraft basically banked sharply to the left and down, as if I had pulled the sticks hard, and went down quite unceremoniously, there was nothing controlled about it, that's why I thought it was perhaps my controller going nuts. Barely had time to try to correct anything, it just went down and out. Never had I seen it do that before. Props were 100% intact before (and after) the "flight". IMU and compass were calibrated. It literally was in the air for less than 10-12 secs before it went nuts. I didn't even have the SD card in yet... so would there even be any flight log? How can I try and see if I can salvage a log from somewhere? The controller perhaps?
Still doesn't sound like an automated response from the drone due to being in a restricted area. Did you have GPS lock before take off? If you did and were in restricted airspace it shouldn't have taken off. If not, and it achieved it in the air, I expect you would have received a warning and it should have just landed.

Did you update your DJI Go app or precise fly database recently? What class airspace were you in?
 
Still doesn't sound like an automated response from the drone due to being in a restricted area
Agreed. It sounds more like an issue that occurs after taking off near a magnetic metal object.
 
Sorry about the crash and hassel, I pretty much have to get an unlock every flight but I have like 50+ waivers for my area (DFW + most of TX). As stated before, if youre in a NFZ the drone simply wont take off. However, if your in atti mode before a good GPS lock and then it locates itself in a NFZ it should hover then auto land. Good luck
 
  • Like
Reactions: combatplane57
OP, do you have any pictures of the damage?

Not yet, haven't had time...Sorry to post and disappear, been a bit swamped this week (gave my 2 week's notice Friday :cool:) and it's been rather crazy at work lol... Anyhow, doing a second glance shows the damage wasn't much fortunately, looks like the "foot" just got more "bent" than cracked at the point where it screws to the folding arm, though I def need to do a more thorough check to make sure everything's kosher. I think the arm folded backwards when it hit the side of the house and that took most of the impact.

I'll have to do some "forensics" on my "controller" phone via USB to see if I can pull any data logs... I was panicking and like an idiot uninstalled and reinstalled the DJI app to try to see if that was the culprit and that might have deleted those last couple flights, I can't seem to find the one where it actually crashed now via the app UI. I think I need an actual dedicated tablet, and some my old backup cell phone to control my drone. I'm always running outta storage space on that phone anyway.

Lots of good info here though guys, I definitely appreciate the help
 
  • Like
Reactions: Former Member
Agreed. It sounds more like an issue that occurs after taking off near a magnetic metal object.


Holy crap, I think you might be right on the money there. I recall I took off from a metallic patio table...
That may have totally messed up the compass or some other sensor?
Weird thing is though, Ive taken off from that patio table before without issues or any compass/sensor warnings...
 
Holy crap, I think you might be right on the money there. I recall I took off from a metallic patio table...
That may have totally messed up the compass or some other sensor?
Weird thing is though, Ive taken off from that patio table before without issues or any compass/sensor warnings...

You can take off from anywhere, even if you are getting a compass error. Once you get a little altitude the error message usually goes away. Usually means exactly that.
 
Holy crap, I think you might be right on the money there. I recall I took off from a metallic patio table...
That may have totally messed up the compass or some other sensor?
Weird thing is though, Ive taken off from that patio table before without issues or any compass/sensor warnings...

You often won't see a compass error at the takeoff point - all that happens is that the local magnetic field (in this case possibly due to magnetized metal parts in the table) affects the compass, so that it gets the aircraft heading wrong and incorrectly initializes the IMU yaw value. When it takes off and leaves the magnetic field of the table, the compass then reads correctly but the IMU yaw value and compass disagree - that's when the compass error typically occurs. And that can cause uncontrolled flight.

You can take off from anywhere, even if you are getting a compass error. Once you get a little altitude the error message usually goes away. Usually means exactly that.

I've rarely seen a compass error on the ground that goes away after takeoff. Compass errors on the ground are generally because the aircraft magnetic field has changed and it needs recalibration or demagnetization, it is in the presence of an external magnetic field that exceeds its limits or, with the M2, because it has moved more than 50 km or it is more than 30 days since the last calibration. Generally it will not start motors with those kinds of errors but, if it does and can take off, the error will not go away.
 
I've rarely seen a compass error on the ground that goes away after takeoff. Compass errors on the ground are generally because the aircraft magnetic field has changed and it needs recalibration or demagnetization, it is in the presence of an external magnetic field that exceeds its limits or, with the M2, because it has moved more than 50 km or it is more than 30 days since the last calibration. Generally it will not start motors with those kinds of errors but, if it does and can take off, the error will not go away.

Actually, I see it almost every day. If a person takes off in a magnetic field that is interfering with the compass, almost every time within 1 foot off the ground, the magnetic interference goes away, and the error message goes away. It will allow you to spool the motors up every single time.

I have learned that the compass calibration message is most times a misnomer. Compass calibration is not necessary, but it is compass interference that actually the case. As you know, temporary compass interference and compass calibration are 2 different things. Interference requires just moving to a different spot, or get out of the spot with interference. That includes moving or climbing vertically or horizontally sometimes as little as 1 foot to escape the interference.
 
You often won't see a compass error at the takeoff point - all that happens is that the local magnetic field (in this case possibly due to magnetized metal parts in the table) affects the compass, so that it gets the aircraft heading wrong and incorrectly initializes the IMU yaw value. When it takes off and leaves the magnetic field of the table, the compass then reads correctly but the IMU yaw value and compass disagree - that's when the compass error typically occurs. And that can cause uncontrolled flight.



I've rarely seen a compass error on the ground that goes away after takeoff. Compass errors on the ground are generally because the aircraft magnetic field has changed and it needs recalibration or demagnetization, it is in the presence of an external magnetic field that exceeds its limits or, with the M2, because it has moved more than 50 km or it is more than 30 days since the last calibration. Generally it will not start motors with those kinds of errors but, if it does and can take off, the error will not go away.
To be clear, I am talking about a compass error message displayed in DJI Go 4 on the top banner. This error message, again, is not representative of an actual compass error, it is a compass disparity. Temporary in nature, due to taking off in an area that is causing this error message does not mean that a compass calibration is needed. It just means that your location has a magnetic disturbance. I have been experimenting with this theory for a few weeks and gave proven myself right every time. There has not been a single issue after takeoff in nearly 20 flights.
 
Actually, I see it almost every day. If a person takes off in a magnetic field that is interfering with the compass, almost every time within 1 foot off the ground, the magnetic interference goes away, and the error message goes away. It will allow you to spool the motors up every single time.

In all the flight logs I've reviewed with compass issues, which is many hundreds now, I've never seen that sequence of events, and it would be completely inconsistent with how the FC uses the compass and IMU. Taking off from a magnetically distorted location is an almost certain recipe for compass/yaw errors in flight, with all the accompanying uncontrolled flight problems that you have no doubt seen in dozens of flight log analysis of events here. If you have a DAT file showing that not happening then I'd certainly like to see it.

I have learned that the compass calibration message is most times a misnomer. Compass calibration is not necessary, but it is compass interference that actually the case. As you know, temporary compass interference and compass calibration are 2 different things. Interference requires just moving to a different spot, or get out of the spot with interference. That includes moving or climbing vertically or horizontally sometimes as little as 1 foot to escape the interference.

It can certainly be a misnomer, as has been explained many times here, but it depends on the model of Mavic. Compass interference (move or calibrate) messages on the ground arise from one of two reasons: either the magnetometers are reporting values that are outside the accepted range (could be external or internal magnetic fields) or, in the case of the Mavic 2, the aircraft is requesting calibration based on time or distance, even though the magnetometers are probably measuring well within the expected range.

If it is a time/distance triggered M2 message then it should still fly just fine, but my M2 won't start motors until it is recalibrated in that situation.

If it is magnetic interference then moving the aircraft away from the source is all that's needed, but the IMU has to be reinitialized otherwise compass errors will occur after takeoff. That will happen spontaneously given time, but it's safer just to restart the aircraft. Taking off is a terrible idea as the means of moving out of the interference, because the IMU yaw value will definitely be incorrect on takeoff, and IMU/compass disagreement will inevitably occur as soon as the aircraft climbs out of the distorted magnetic field.

So I'm not aware of any compass interference error on the ground that would allow you to takeoff, go away after takeoff, and not lead to serious IMU/compass disagreements afterwards.

This topic has been explained and discussed ad nauseam by me, @BudWalker, @Meta4 and others, supported by numerous sets of data, so I'm very surprised to see you propose an explanation and strategy that is completely contradictory with what I would say is the established explanation.
 
Last edited:
  • Like
Reactions: msinger
To be clear, I am talking about a compass error message displayed in DJI Go 4 on the top banner. This error message, again, is not representative of an actual compass error, it is a compass disparity. Temporary in nature, due to taking off in an area that is causing this error message does not mean that a compass calibration is needed. It just means that your location has a magnetic disturbance. I have been experimenting with this theory for a few weeks and gave proven myself right every time. There has not been a single issue after takeoff in nearly 20 flights.

Interestingly @BudWalker and I have discussed mitigation strategies for IMU/compass disagreement once airborne in this situation, and it's not clear why the FC is not programmed to trust the compass and reset the IMU yaw in this situation. In most cases that would fix the problem, and while that could conceivably cause issues if the aircraft flew into a distorted magnetic field, that scenario is much less common than starting up in a distorted field. I've not seen any evidence that any of the latest firmware has changed in this respect, but I haven't yet seen any cases of that scenario with the latest M2 firmware so I can't rule out that they haven't changed it, although that would be a very significant change. Which platform and firmware version are you experimenting with?

Unless you have examined the DAT files and confirmed magnetic distortion on the ground that then goes away on takeoff without any issues, I'm going to have to say that you are probably mistaken in your interpretation of what you are seeing.
 
  • Like
Reactions: msinger
Here is a frequent scenario. I turn on a drone. I get a compass calibration error message from DJI Go 4. I pick up the drone and make sure that the orientation of the drone on the DJIG4 app map is correct and yaws when I yaw the drone in my hand. I put the drone back down on the ground. Still get a compass error message. I disregard it, and start an autotakeoff, making sure I select the precision landing box. After the drone climbs to its preset autotakofff height, the compass calibration error message goes away, and never see a sign of it the whole rest of the flight.

I have done this countless times, and have not had the toilet bowl effect even one single time.
 
Unless you have examined the DAT files and confirmed magnetic distortion on the ground that then goes away on takeoff without any issues, I'm going to have to say that you are probably mistaken in your interpretation of what you are seeing.

Sorry but I dont agree with that logic. My interpretation is spot on. Its DJI Gos interpretation of what the drone is "sensing" that is mistaken and throwing off an ill worded error message.
 
Here is a frequent scenario. I turn on a drone. I get a compass calibration error message from DJI Go 4. I pick up the drone and make sure that the orientation of the drone on the DJIG4 app map is correct and yaws when I yaw the drone in my hand. I put the drone back down on the ground. Still get a compass error message. I disregard it, and start an autotakeoff, making sure I select the precision landing box. After the drone climbs to its preset autotakofff height, the compass calibration error message goes away, and never see a sign of it the whole rest of the flight.

I have done this countless times, and have not had the toilet bowl effect even one single time.

Right - but that's completely different - there is no magnetic interference in that scenario. That's confirmed by the fact that the arrow is pointing in the correct direction and yaws properly as you turn it. That's precisely why that is the recommended check for magnetic interference. And that's also why you get no compass yaw errors in flight - the compass correctly initialized the IMU.
 
Sorry but I dont agree with that logic. My interpretation is spot on. Its DJI Gos interpretation of what the drone is "sensing" that is mistaken and throwing off an ill worded error message.

OK - if you prefer to make guesses as to what was happening than consult the DAT file to see what actually happened then there is little point continuing this discussion.
 
Right - but that's completely different - there is no magnetic interference in that scenario. That's confirmed by the fact that the arrow is pointing in the correct direction and yaws properly as you turn it. That's precisely why that is the recommended check for magnetic interference. And that's also why you get no compass yaw errors in flight - the compass correctly initialized the IMU.

Yes but why does it still show a compass error message in DJI Go4?
 
OK - if you prefer to make guesses as to what was happening than consult the DAT file to see what actually happened then there is little point continuing this discussion.

I dont have to consult the Dat file to know that once the drone gets off the ground, the error message goes away.
 
That's what you need to check the DAT file to find out. Without the data it's just speculation.

Does the "Compass error, move aircraft or calibrate compass" error message that DJIG4 shows mean that the compass is being interfered with even though the map is accurate? Can the map be accurate, but the above error message will still appear?

Another thought. Lets say an owner decides to always take off from the same spt and gets this error message every time. The drone allows him to take off, and within a foot of the ground. Lets further say that the message is due to magnetic interference. Is it possible that a long term consequence could be permanent damage to the compass?
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
131,115
Messages
1,559,955
Members
160,090
Latest member
Electrakill21